Docstoc

Photocharge Transport and Recombination Measurements in

Document Sample
Photocharge Transport and Recombination Measurements in Powered By Docstoc
					December 2002            •      NREL/SR-520-33173




Photocharge Transport and
Recombination Measurements
in Amorphous Silicon Films and
Solar Cells by Photoconductive
Frequency Mixing
Final Subcontract Report
20 April 1998–30 June 2002

R. Braunstein, M. Boshta, S. Sheng,
A. Kattwinkel, J. Liebe, and G. Sun
University of California
Los Angeles, California




          National Renewable Energy Laboratory
          1617 Cole Boulevard
          Golden, Colorado 80401-3393
          NREL is a U.S. Department of Energy Laboratory
          Operated by Midwest Research Institute • Battelle • Bechtel
          Contract No. DE-AC36-99-GO10337
December 2002                 •      NREL/SR-520-33173




Photocharge Transport and
Recombination Measurements
in Amorphous Silicon Films and
Solar Cells by Photoconductive
Frequency Mixing
Final Subcontract Report
20 April 1998–30 June 2002

R. Braunstein, M. Boshta, S. Sheng,
A. Kattwinkel, J. Liebe, and G. Sun
University of California
Los Angeles, California




NREL Technical Monitor: B. von Roedern
Prepared under Subcontract No. XAK-8-17619-24




             National Renewable Energy Laboratory
             1617 Cole Boulevard
             Golden, Colorado 80401-3393
             NREL is a U.S. Department of Energy Laboratory
             Operated by Midwest Research Institute • Battelle • Bechtel
             Contract No. DE-AC36-99-GO10337
                                                       NOTICE

This report was prepared as an account of work sponsored by an agency of the United States
government. Neither the United States government nor any agency thereof, nor any of their employees,
makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,
completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents
that its use would not infringe privately owned rights. Reference herein to any specific commercial
product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily
constitute or imply its endorsement, recommendation, or favoring by the United States government or any
agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect
those of the United States government or any agency thereof.


                           Available electronically at http://www.osti.gov/bridge

                           Available for a processing fee to U.S. Department of Energy
                           and its contractors, in paper, from:
                                    U.S. Department of Energy
                                    Office of Scientific and Technical Information
                                    P.O. Box 62
                                    Oak Ridge, TN 37831-0062
                                    phone: 865.576.8401
                                    fax: 865.576.5728
                                    email: reports@adonis.osti.gov

                           Available for sale to the public, in paper, from:
                                   U.S. Department of Commerce
                                   National Technical Information Service
                                   5285 Port Royal Road
                                   Springfield, VA 22161
                                   phone: 800.553.6847
                                   fax: 703.605.6900
                                   email: orders@ntis.fedworld.gov
                                   online ordering: http://www.ntis.gov/ordering.htm



      Printed on paper containing at least 50% wastepaper, including 20% postconsumer waste
                                           Preface

The prime candidate material for thin-film photovoltaic high efficient solar cells for large-scale
power generation is hydrogenated amorphous silicon and alloys. The objectives of the
technology in this field are to achieve stable and efficient units for cost effective bulk-power
generation. The strategy in this field is to optimize amorphous thin film growth for greater
efficiency, to suppress the light-induced instability, and to achieve high deposition rates so as to
improve the throughput for a given machine, and also to reduce the costs and the capital
investment. Material preparation efforts of amorphous semiconductors have concentrated on the
reduction of "Urbach" edges, sub-bandgap absorption, and the density of deep defects to the end
to maximize the photoconductive gain of the material. Most material efforts have been to
optimize mobility-lifetime product (µτ) as measured by steady state photoconductivity which
does not determine µ and τ separately. To evaluate various photocharge transport models, it is
essential that a simultaneous determination of the mobility and lifetime be performed so as to
predict the performance of solar cells. We have developed a photomixing technique to separately
determine the mobility and lifetime to characterize materials to predict solar cell performance
and to allow the testing of new materials and devices in actual solar cell configurations. The
present program formed part of the NREL High-Bandgap Alloy Team, the Metastability and the
Mid- and Low-Bandgap Alloy Teams. Various groups were concerned with material synthesis
and device fabrication. The UCLA Group performed photoconductive frequency mixing
measurements on these material and solar cell devices to determine the optimum growth
conditions for photocharge transport. The continuous feedback of the results of the UCLA Group
to the synthesis groups relating material properties to device performance gave insight into the
light-induced degradation mechanisms.




                                                 i
                                                     Table of Contents

Preface.............................................................................................................................................. i
Table of Contents............................................................................................................................ ii
List of Figures ................................................................................................................................ iv
List of Tables .................................................................................................................................. x
Executive Summary ........................................................................................................................ 1
Introduction..................................................................................................................................... 3
Results and Analysis ....................................................................................................................... 4
    1.     Improvement of Photomixing Instrumentation................................................................... 4

    2.     Initial experiments on single film samples with coplanar and perpendicular
           contact geometry................................................................................................................. 6

    3.     Hydrostatic pressure dependence of the charge transport in amorphous silicon ................ 9

    4.     Hydrostatic Pressure Dependence of Charged Transport and Small Angle X-ray
           Scattering Measurements .................................................................................................. 14

    5.     The “uninterrupted growth/annealing” method ................................................................ 15

    6.     Charge transport properties near and above the transition from amorphous to
           microcrystalline silicon..................................................................................................... 22
           6.1         Material produced at NREL.................................................................................. 22
           6.2         Charge transport properties near and above the transition from amorphous to
                       microcrystalline silicon produced at MVSystems ................................................ 32
           6.3         Charge transport properties of microcrystalline silicon prepared by Pulsed
                       PECVD technique (MVSystems Inc.) .................................................................. 47
    7.     High Deposition Rate Preparation of a-Si:H by HWCVD ............................................... 54
           7.1         First series of samples........................................................................................... 54
           7.2         The second series of samples................................................................................ 63
           7.3         Characterization of high deposition rate HWCVD a-Si:H films deposited
                       on stainless steel substrates by Driven Level Capacitance Spectroscopy and
                       transient photocapacitance at the University of Oregon. ...................................... 75
           7.4         The effect of deposition rate on the transport properties of HWCVD
                       a-Si:H films with respect to the substrate temperature, deposition pressure
                       and silane flow rate ............................................................................................... 77
    8.     Measurements of amorphous (Si,Ge) alloys ..................................................................... 85
           8.1         Measurements of homogenous a-SiGe alloy samples produced at NREL ........... 85
           8.2         Charge transport properties of PECVD a-SiGe:H films produced by BP Solar... 93


                                                                          ii
      8.3        Electrical and optical properties of high quality low bandgap
                 amorphous (Ge, Si) alloys prepared by ECR plasma deposition.......................... 96
9.    Attempt at finding evidence of the existence of long range potential
      fluctuations in single crystal GeSi alloys........................................................................ 103

10. Photoconductive Frequency Mixing Measurements on TCO......................................... 104

11. Comparison of intrinsic film properties and device performance .................................. 105

12. Photomixing Measurements on SiGe P-I-N Devices (Supplied by
    Xunming Deng, Univerity of Toledo) ............................................................................ 110

13. Photo-emission in Air ..................................................................................................... 119
      I.         Statement of the problem .................................................................................... 119
      II.        Implementation of photo-emission in Air........................................................... 119
      III.       Results................................................................................................................. 120
14. Subcontract Supported Publications ............................................................................... 134

15. References....................................................................................................................... 135




                                                                  iii
                                                      List of Figures

Figure 1.        Current through the µc-Si sample MVS723 (supplied by MVSystems,
                 see previous section) during one 50ms BIAS pulse under illumination (a)
                 and in dark condition (b). ............................................................................................ 4
Figure 2.        Photomixing and dc – signal averaging and fitting (solid curves) .............................. 5
Figure 3.        Sample schematics for measurements in two field directions..................................... 6
Figure 4.        DC and AC currents with perpendicular contact geometry(Film thickness 5244Å) .. 7
Figure 5.        Field dependent mixing power for both configurations. For the TCO–a-Si:H–Al
                 sample, the zero-field point was shifted according to the mixing power
                 minimum at V = -0.5V (Fig. 4) ................................................................................... 8
Figure 6.        The Photoconductivity vs. Illumination time for three different pressures............... 10
Figure 7.        The drift mobility and lifetime as functions of illumination time under different
                 pressures. ................................................................................................................... 10
Figure 8.  Infrared transmission of T516 under annealed (“ann”), pressured (“press”), and
           reannealed (“reann”) conditions................................................................................ 12
Figure 9. T516 – film thickness after pressure application....................................................... 12
Figure 10. Pressure-dependent transport parameters. After initial annealing, the sample was
           pressurized and released again. We performed four cycles in the same way, with
           the transport parameters monitored in situ. The x-axis position is associated with
           the over-all illumination period during the measurement. ........................................ 12
Figure 11. DC photocurrent and photomixing signal under different pressures in single
           crystalline silicon. Both decrease with increasing pressure. The curves show only
           an elastic effect. ......................................................................................................... 13
Figure 12. Samples G202 and G179, dilution ratio is kept constant at 1:5 ................................ 17
Figure 13. Samples G203 and G204, dilution ratio is kept constant at 1:5 ................................ 18
Figure 14. Samples I70426 and I80929. ..................................................................................... 19
Figure 15. Samples P60715I with flat and rough surface. These are prepared by employing
           the “uninterrupted growing/annealing” technique. .................................................. 20
Figure 16. Photoconductivity of amorphous and microcrystalline silicon vs hydrogen content 24
Figure 17. The photomixing lifetime of amorphous and microcrystalline Si:H vs hydrogen
           content ....................................................................................................................... 25
Figure 18. The mobility of amorphous and microcrystalline Si:H vs. hydrogen content........... 25
Figure 19. Drift mobility in the case of entirely amorphous samples......................................... 26
Figures 20 and 21. Range and depth of long-range potential fluctuations in amorphous
            and microcrystalline silicon-hydrogen ...................................................................... 26




                                                                      iv
Figure 22. Normalized photoconductivity vs. illumination duration for T516, T531and
           T534 (amorphous state). The light intensities are given in the Figure...................... 27
Figure 23. Photoconductivity, mobility and lifetime vs. illumination duration, T529. .............. 28
Figure 24. Photoconductivity, mobility and lifetime vs. illumination duration, T530. .............. 28
Figure 25. Photoconductivity, mobility and lifetime vs. illumination duration, T532. .............. 29
Figure 26. IR – spectra, sample T516 (a-Si:H) on polycrystalline substrate.............................. 30
Figure 27. Stretching modes in a-Si:H. According to this double peak, a significant part
           of H atoms is bonded as monohydide in a dense Si network. ................................... 31
Figure 28. Stretching modes in microcrystalline silicon – high dilution (note the low 2000 -
           absorption)................................................................................................................. 31
Figure 29. Bending mode related absorption peak at around 900cm-1 in the mixed state
           sample T532. ............................................................................................................. 31
Figure 30. Microcrystalline Silicon with low dilution ratio ....................................................... 31
Figure 31. Deconvolution of 2000/2100-peak (T516)................................................................ 32
Figure 32. Stretching modes for several amorphous and microcrystalline samples................... 32
Figure 33. MVS722 (a-Si:H) Results for the mobility, lifetime, and photoconductivity
           during light-soaking. ................................................................................................. 33
Figure 34. MVS823 (a-Si:H) Results for the mobility, lifetime, and photoconductivity
           during light-soaking. ................................................................................................. 33
Figure 35. MVS723 (µc-Si) Results for the mobility, lifetime, and photoconductivity
           during light-soaking .................................................................................................. 34
Figure 36. Light-induced changes of photoconductivity, mobility, and lifetime for
           sample MVS 968 ....................................................................................................... 36
Figure 37. XRD patterns of the HWCVD intrinsic Si films samples. The H-dilution
           degree and film thickness are also given................................................................... 40
Figure 38. The photoconductivity of the samples as a function of H-dilution ........................... 41
Figure 39. The lifetime (a) and mobility (b) as a function of H-dilution ................................... 41
Figure 40. The depth (a) and range (b) of potential fluctuations as a function of H-dilution .... 42
Figure 41. The relative change in the density of charged defects as a function of H-dilution ... 42
Figure 42. The photoconductivity of the samples as a function of H-dilution ........................... 44
Figure 43. The mobility (a) and lifetime (b) as a function of H-dilution ................................... 45
Figure 44. The depth (a) and range (b) of potential fluctuations as a function of H-dilution .... 45
Figure 45. The relative change in the density of charged defects as a function of H-dilution ... 46
Figure 46. The effect of H-dilution on the photoconductivity of the samples............................ 49
Figure 47. The effect of H-dilution on the mobility (a) and lifetime (b) .................................... 50



                                                                    v
Figure 48. The effect of H-dilution on the depth (a) and range (b) of potential fluctuations .... 50
Figure 49. The effect of H-dilution on the relative change in the density of charged defects.... 51
Figure 50. The effect of substrate temperature on the photoconductivity of the samples .......... 52
Figure 51. The effect of substrate temperature on the mobility (a) and lifetime (b) .................. 52
Figure 52. The effect of substrate temperature on the depth (a) and range (b) of potential
           fluctuations ................................................................................................................ 53
Figure 53. The effect of substrate temperature on the relative change in the density
           of charged defects...................................................................................................... 53
Figure 54. The photoconductivity as a function of deposition rate ............................................ 56
Figure 55. The drift mobility as a function of deposition rate.................................................... 56
Figure 56. The lifetime as a function of deposition rate ............................................................. 56
Figure 57. Field dependence of the drift mobility (a) and lifetime (b) for sample L183 ........... 57
Figure 58. The depth (a) and range (b) of the long range potential fluctuations as a
           function of deposition rate......................................................................................... 57
Figure 59. The relative change in the charged defects as a function of deposition rate ............. 58
Figure 60. The hydrogen content as a function of deposition rate ............................................. 58
Figure 61. Normalized photoconductivity, drift mobility and lifetime for high deposition
           rate HWCVD a-Si:H samples as a function of illumination ..................................... 59
Figure 62. The light-induced changes in the photoconductivity, drift mobility and lifetime
           as a function of deposition rate.................................................................................. 60
Figure 63. The light-induced changes in the depth and range of potential fluctuations as a
           function of deposition rate......................................................................................... 61
Figure 64. The light-induced relative change in the charged defect density as a function
           of deposition rate. ...................................................................................................... 62
Figure 65. The photoconductivity as a function of deposition rate ................................................
Figure 66. The lifetime as a function of deposition rate ............................................................. 65
Figure 67. The drift mobility as a function of deposition rate.................................................... 65
Figure 68. Field dependence of the drift mobility (a) and the lifetime (b) as a function
           of deposition rate ....................................................................................................... 66
Figure 69. The depth (a) and range (b) of the long range potential fluctuations as a
           function of deposition rate......................................................................................... 67
Figure 70. The relative change in the density of charged defects as a function of
           deposition rate ........................................................................................................... 67
Figure 71. The light-induced changes in the photoconductivity as function of deposition rate. 70
Figure 72. The light-induced changes in the drift mobility as a function of deposition rate...... 70
Figure 73. The light-induced changes in the lifetime as a function of deposition rate.. ……….70


                                                                    vi
Figure 74. Effect of light-soaking on the electric field dependence of the drift mobility
           as a function of deposition rate.................................................................................. 71
Figure 75. Effect of light-soaking on the electric field dependence of the lifetime as
           a function of deposition rate...................................................................................... 72
Figure 76. The light-induced changes in the depth (a) and range (b) of potential
           fluctuations as a function of deposition rate.............................................................. 73
Figure 77. The light-induced relative change in the charged defect density as a function
           of deposition rate ....................................................................................................... 74
Figure 78. Deposition pressure dependence of the conductivity of the samples as
           a function of deposition rate...................................................................................... 78
Figure 79. Deposition pressure dependence of the mobility (a) and lifetime (b) as
           a function of deposition rate...................................................................................... 79
Figure 80. Deposition pressure dependence of the depth (a) and range (b) of the
           potential fluctuations as a function of deposition rate............................................... 79
Figure 81. Deposition pressure dependence of the relative change in the density
           of charged defects as a function of deposition rate ................................................... 80
Figure 82. Deposition pressure dependence of the hydrogen content as a function
           of deposition rate ....................................................................................................... 80
Figure 83. Silane flow rate dependence of the conductivity of the samples as a function
           of deposition rate ....................................................................................................... 80
Figure 84. Silane flow rate dependence of the mobility (a) and lifetime (b) as a function
           of deposition rate ....................................................................................................... 81
Figure 85. Silane flow rate dependence of the depth (a) and range (b) of the potential
           fluctuations as a function of deposition rate.............................................................. 81
Figure 86. Silane flow rate dependence of the relative change in the density of charged
           defects as a function of deposition rate ..................................................................... 82
Figure 87. Silane flow rate dependence of the hydrogen content as a function of
           deposition rate ........................................................................................................... 82
Figure 88. Substrate temperature dependence of the conductivity of the samples as
           a function of deposition rate...................................................................................... 83
Figure 89. Substrate temperature dependence of the mobility (a) and lifetime (b) as
           a function of deposition rate...................................................................................... 83
Figure 90. Substrate temperature dependence of the depth (a) and range (b) of the
           potential fluctuations as a function of deposition rate............................................... 84
Figure 91. Substrate temperature dependence of the relative change in the density
           of charged defects as a function of deposition rate ................................................... 84
Figure 92. Substrate temperature dependence of the hydrogen content as a function
           of deposition rate ....................................................................................................... 84



                                                                  vii
Figure 93. The normalized photoconductivity, mobility, and lifetime in a-Si:H as a
           function of illumination time..................................................................................... 86
Figure 94. The normalized photoconductivity, mobility, and lifetime as a function of
           illumination time. The GeH4 gas flow ratio is 3%. ................................................... 86
Figure 95. The normalized photoconductivity, mobility, and lifetime as a function of
           illumination time. The GeH4 gas flow ratio is 3%. ................................................... 87
Figure 96. The normalized photoconductivity, mobility, and lifetime as a function of
           illumination time. The GeH4 gas flow ratio is 8%. ................................................... 87
Figure 97. The normalized photoconductivity, mobility, and lifetime as a function of
           illumination time. The GeH4 gas flow ratio is 17%. ................................................. 88
Figure 98. RMS roughness vs. GeH4 gas ratio. .......................................................................... 89
Figure 99. Photoconductivity, mobility, lifetime, LRPF range and depth vs.
           germanium hydrogen gas flow ratio (annealed state). .............................................. 90
Figures 100-102. The AFM scanning images and surface height distribution of the
           samples HGe 121, 113, and 100................................................................................ 91
Figures 103-104. The AFM scanning images and surface height distribution of the
           samples HGe 114 and HGe 118 ................................................................................ 92
Figure 105. The photoconductivity of the samples as a function of Ge content .......................... 94
Figure 106. The mobility (a) and lifetime (b) of the samples as a function of Ge content........... 94
Figure 107. The depth (a) and range (b) of potential fluctuations as a function of Ge content.... 95
Figure 108. The relative change in the density of charged defects as a function of Ge content .. 95
Figure 109. The optical absorption spectra of the samples. The Si content and Tauc gapEg ...... 99
Figure 110. The sub-gap absorption spectra of some of the samples. The Si content
            and the Urbach energy (Eo) of the valence band-tail are also given ....................... 100
Figure 111. The photoconductivity of the samples as a function of the Tauc gap ..................... 100
Figure 112. The mobility (a) and lifetime (b) of the samples as a function of the Tauc gap ..... 101
Figure 113. The electric field dependence of the drift mobility as a function of the
            Tauc gap. The solid curves are fit of the data to equation (1)................................. 101
Figure 114. The depth (a) and range (b) of potential fluctuations as a function of the
            Tauc gap .................................................................................................................. 102
Figure 115. The relative change in the density of charged defects as a function of the
            Tauc gap. ................................................................................................................. 102
Figure 116. Light-induced changes in photoconductivity, mobility, and lifetime
            of film R8796 (H2 diluted, using ............................................................................. 106
Figure 117. Light-induced changes in photoconductivity, mobility, and lifetime
            of film R8794 (H2 diluted, using ............................................................................ 106
Figure 118. dc-photoconductivity for nip-devices in the annealed state .................................... 108


                                                                   viii
Figure 119. Square root of mixing signal for nip-devices in the annealed state......................... 108
Figure 120. Normalized decay of the dc photocurrent in short circuit condition
            and the back-bias mixing signal in nip device R8791............................................. 109
Figure 121. Normalized decay of the dc photocurrent in short circuit condition
            and the back-bias mixing signal in nip device R8792............................................. 109
Figure 122. Normalized decay of the dc photocurrent in short circuit condition
            and the back-bias mixing signal in nip device R8793............................................. 110
Figure 123. DC photocurrent and the square root of the photomixing power
            (inset: derivative of sqrt(Pmix)). ............................................................................... 112
Figure 124. Light intensity dependent photomixing current. Right hand-side:.......................... 113
Figure 125. dc and mixing currents after different illumination times
            (Sample Toledo-GD112)......................................................................................... 114
Figure 126. dc and mixing currents after different illumination times
            (Sample Toledo-GD111)......................................................................................... 115
Figure 127. Sample Toledo-GD110............................................................................................ 116
Figure 128. dc and mixing currents after different illumination times
            (Sample Toledo-GD109)......................................................................................... 117
Figure 129. BIAS-dependent ac – photocurrent for two different spots .................................... 118
Figure 130. Schematic diagram of Photo-emission in air........................................................... 121
Figure 131. Electron energies between the anode and cathode in a diode under bias conditions.
            (a) Anode reversed bias (retarding field); (b) emission of space charge limited barrier
            results from space- charge just outside cathode surface; (c) saturation emission ... 122
Figure 132. Photo-emission current vs voltage bias ................................................................... 123
Figure 133. The photo-emission intensity of ITO and a-Si:H on stainless steel substrate ......... 124
Figure 134. The photo-emission intensity of ITO on a-Si:H on stainless steel substrate.......... 125
Figure 135. The photo-emission intensity of ITO/pin/SS layers. .............................................. 126
Figure 136. The photo-emission intensity of Pd/ni/SS layers. ................................................... 127
Figure 137. The photo-emission intensity of a-Si:H on c-Si substrate. ...................................... 128
Figure 138. The photo-emission intensity the 16 pads which are a-Si:H to µc-Si:H
            and a-SiGe:H to µc-SiGe:H on stainless steel substrate. ........................................ 129
Figure 139(a). The photo-emission intensity of a-SiC:H on stainless steel substrate. .............. 130
Figure 139(b). The photo-emission intensity of a-SiC:H on stainless steel substrate. .............. 131
Figure 139(c). The photo-emission intensity of a-SiC:H on stainless steel substrate. .............. 132
Figure 139(d). The photo-emission intensity of a-SiC:H on stainless steel substrate. .............. 133




                                                               ix
                                               List of Tables
Table 1.    Deposition conditions .............................................................................................. 16
Table 2.    Long range potential fluctuations ............................................................................ 21
Table 3.    Summary of the sample characterization results (Qi Wang). .................................. 23
Table 4.    LRPF Range and Depth for the MVS – samples (the thickness of the
            intrinsic layers were about 5000Å). ......................................................................... 34
Table 5.    Sample data as provided by MVSystems................................................................. 35
Table 6.    Preparation conditions and thickness of HWCVD µc-Si:H samples provided
            by MVSystems Inc................................................................................................... 37
Table 7.    The effect of hydrogen dilution on the transport properties of µc-Si:H samples
            provided by MVSystems Inc. .................................................................................. 38
Table 8.    The effect of substrate temperature on the transport properties of µc-Si:H
            samples provided by MVSystems Inc...................................................................... 38
Table 9.    The effect of flament-substrate distance on the transport properties of
            µc-Si:H samples provided by MVSystems Inc........................................................ 39
Table 10.   Preparation conditions and thickness of HWCVD µc-Si:H samples provided
            by MVSystems Inc................................................................................................... 43
Table 11.   Deposition conditions and properties of Pulsed PECVD µc-Si:H samples
            provided by MVSystems.......................................................................................... 48
Table 12.   XRD characterizations of Pulsed PECVD µc-Si:H samples provided
            by MVSystems......................................................................................................... 48
Table 13.   Growth conditions and properties of high deposition rate HWCVD samples......... 55
Table 14.   Growth conditions and properties of high deposition rate HWCVD samples......... 64
Table 15.   Characterization of high deposition rate HWCVD a-Si:H samples deposited
            on stainless steel substrates. (supplied by the University of Oregon)...................... 75
Table 16.   Preparation Conditions of High Deposition Rate HWCVD a-Si:H Samples. ......... 78
Table 17.   The characteristics of a-SiGe:H and a-Si:H............................................................. 85
Table 18.   The photoconductivity, drift mobility, lifetime, and range and depth of
            long range potential fluctuations in annealed state for a-SiGe:H alloys
            prepared with different GeH4 gas flow ratio............................................................ 88
Table 19.   The photoconductivity, drift mobility, lifetime, and range and depth of
            long range potential fluctuations in light-soaked state for a-SiGe:H alloys
            prepared with different GeH4 gas flow ratio............................................................ 89
Table 20.   The Ge content and film thickness for the samples produced by BP Solar............. 93
Table 21.   Growth conditions and properties of ECR a-GeSi:H/a-Ge:H samples.................... 97



                                                             x
Table 22.   Growth conditions of the films and the respective i-layers in the n-i-p structures.105
Table 23.   Long-range potential fluctuations. ......................................................................... 107
Table 24.   I-V data for a-SiGe solar cells with different i-layers (~2000Å)........................... 117




                                                          xi
                                 Executive Summary
The instrumentation of the photomixing measurements has been improved by applying BIAS
pulses of arbitrary width and frequency. This pulsed photomixing method enables the
determination of transport properties of samples with high conductivity, and can considerably
improve the accuracy of the measurements at low electrical fields applied.

Measurements of the transport parameters under hydrostatic pressure were initiated to investigate
whether the application of hydrostatic pressure could alter light induced degradation behavior
and as a possible test of the various theories that have been recently been developed to explain
the cause of light-degradation involving the mobility of hydrogen. Furthermore, we investigated
the hydrostatic pressure dependence of small angle X-ray scattering measurements to try to find
out the origin of the inelastic effect observed.

The charge transport and structural properties and their light-induced changes near and above the
transition from amorphous to microcrystalline silicon prepared by HWCVD and Pulsed PECVD
techniques have been investigated in detail by the photomixing, FTIR and XRD techniques. A
study of the effects of the H-dilution, substrate temperature and filament-substrate distance (F-S)
on transport properties of the HWCVD samples showed that higher H-dilution, higher substrate
temperature and shorter F-S can help improve the transport properties. XRD results of the Pulsed
PECVD samples showed that the ratio of the (220) XRD peak area to the (111) peak area is seen
to be in the range of 700% for the films near the transition regime, indicating that the transition
films are strongly (220) oriented. Moreover, higher substrate temperature is required for the
Pulsed PECVD technique to produce high quality films and related devices near the transition
regime. We found both improved stability against light-soaking and dramatically different values
for the lifetime and mobility close to the onset of microcrystallinity as compared to the
amorphous state. In particular, the lifetime of charge carriers in some of the µc-Si:H samples
turns out to lie about two orders of magnitude higher than that of a-Si:H films. The mobility, on
the other hand, is shown rather to decrease in and above the transition regime. Additional
measurements of the range and the depth of long range potential fluctuations yield a possible
explanation for our results in that grain boundaries may serve as scattering centers and barriers
against recombination.

The electronic transport properties in both annealed and light-soaked states of high deposition
rate a-Si:H films prepared by HWCVD has been investigated in detail by the photomixing
technique and Driven Level Capacitance Spectroscopy and transient photocapacitance. The high
deposition rate (up to 1 µm/min.) was achieved by increasing deposition pressure, silane flow
rate, and decreasing filament-to-substrate distance. The effects of the deposition rate on the
resultant film properties with respect to the substrate temperature, deposition pressure and silane
flow rate were studied. It was found that the film transport properties do not change
monotonically with increasing deposition rate. The photoconductivity peaks at ~70-90 Å/s,
where both the drift mobility and lifetime peak, consistent with the deposition rate dependence of
the range and depth of the potential fluctuations. High quality, such as a photoconductivity-to-
dark-conductivity ratio of ~105 and nearly constant low charged defect density, can be
maintained at deposition rates up to ~130 Å/s, beyond which the film properties deteriorate


                                                1
rapidly as a result of an enhanced effect of long-range potential fluctuations due to a
considerable increase in the concentration of the charged defects. Our results indicate that
medium silane flow rate, low pressure, and higher substrate temperature are generally required to
maintain high quality films at high deposition rates.

The charge transport and optical properties of amorphous (Si,Ge) alloys prepared by HWCVD,
PECVD and ECR plasma deposition techniques as a function of alloy composition have been
investigated in detail by the photomixing technique and optical absorption spectroscopy.
Evidence for the presence of long-range potential fluctuations in a-SiGe:H was revealed from the
measurements of electric field dependence of the drift mobility, and the effect of the long-range
potential fluctuations is enhanced by the addition of Ge to the alloy system that results in the
deterioration of the opto-electronic properties of a-SiGe:H. It was found that at a composition of
~10% Ge in Si for HWCVD, and ~20% for PECVD, the photoresponse begins to decrease
monotonically with increasing Ge content due to the decreases in both the drift mobility and the
lifetime as a result of an increase in the concentration of charged defects, which lead to the long-
range potential fluctuations whose depth increases, while the range decreases. On the other hand,
High quality low bandgap a-(Ge, Si):H alloys at the Ge end were successfully prepared using
ECR plasma deposition with high H dilution and ppm B-doping. Incorporating these high quality
materials into devices leads to much lower gap a-(Ge, Si) solar cells (down to ∼1 eV in a-Ge:H)
with acceptable performance. It was found that at SiH4/(SiH4+GeH4) ~30%, the photoresponse
begins to decrease rapidly with increasing Si content due to the decreases in the mobility and
lifetime, and meanwhile, both the charged defect density and the Urbach energy increase
significantly. The latter indicates an increase in the compositional disorder. It is the potential
fluctuations whose effect can be also enhanced by incorporating Si to the alloy system that result
in the deterioration of the opto-electronic properties of a-(Ge, Si):H alloys, similar to the case of
the incorporation of Ge at the Si end. The increased charged scattering centers and compositional
disorder upon adding Si or Ge to the alloys observed play an important role in the potential
fluctuations.

We also attempted to employ the photomixing technique to find evidence of the existence of
long-range potential fluctuations in single crystal GeSi alloys, and to measure the drift mobility
of TCO.

Photomixing experiments were initiated on study of the impact of the changed contact geometry
on the results of our photomixing measurements. Photomixing experiments were also initiated on
p-i-n devices, and on comparison of intrinsic film properties and related device performance.

Time resolved photo- and thermoelectric effects (TTE) were used to simultaneously determine
the thermal diffusivity, carrier lifetime, carrier mobility, and trap level density in crystalline and
amorphous Si (a-Si:H) and Si/Ge (a-Si/Ge:H) samples.




                                                   2
                                      Introduction
The research pursued during the past three years under NREL subcontract #XAK-8-17619-24
were part of a collaboration with members of the NREL Wide-bandgap Alloy Team, the
Metastability and the Mid- and Low-Bandgap Alloy Teams. The tasks were focused on the
characterization of the charge transport, opto-electronic and structural properties of a number of
amorphous and microcrystalline semiconductors prepared by a number of techniques. The
dominant approach to accomplish the tasks of the present phase of the program is the
photoconductive frequency mixing technique. This technique enabled us to separately determine
the drift mobility and the photomixing lifetime of the photogenerated carriers [1-6]. The
technique is based on the idea of heterodyne detection for photoconductors. When two similarly
polarized monochromatic optical beams of slightly different frequencies are incident upon a
photoconductor, the photocurrent produced, when a dc bias applied, will contain components
resulting from the square of the sum of the incident electric fields. Consequently, a photocurrent
composed of a dc and a microwave current due to the beat frequency of the incident fields will
be produced; these two currents allow a separate determination of the drift mobility and the
photomixing lifetime. In the present work, we improved the instrumentation of the photomixing
measurements by applying BIAS pulses of arbitrary width and frequency. The longitudinal
modes of a He-Ne laser were employed to generate a beat frequency of 252 MHz; all the
measurements were performed at this frequency for the data indicated in the accompanying
figures and tables. Employing this technique, as well as other techniques, including FTIR, XRD,
SAXS, Optical Spectroscopy, etc., the following topics were explored whose results will be
presented in the following sections:

   1. Improvement of Photomixing Instrumentation.
   2. Initial experiments on single film samples with coplanar and perpendicular contact
       geometry.
   3. Hydrostatic pressure dependence of the charge transport in amorphous silicon.
   4. Hydrostatic pressure dependence of charged transport and Small Angle X-ray Scattering
       measurements.
   5. Photomixing measurements on samples prepared by the “uninterrupted
       growth/annealing” method.
   6. Charge transport and structural properties near and above the transition from amorphous
       to microcrystalline silicon prepared by HWCVD and Pulsed PECVD.
   7. Charge transport properties of high deposition rate HWCVD a-Si:H.
   8. The charge transport and optical properties of amorphous (Si,Ge) alloys prepared by
       HWCVD, PECVD and ECR plasma deposition.
   9. Attempt at finding evidence of the existence of long range potential fluctuations in single
       crystal GeSi alloys.
   10. Photoconductive frequency mixing measurements on TCO.
   11. Comparison of intrinsic film properties and device performance.
   12. Photomixing measurements on SiGe P-I-N Devices.




                                                3
                                                  Results and Analysis

1. Improvement of Photomixing Instrumentation

Lately, several groups have taken steps into development and investigation of silicon films in the
microcrystalline or µc-/a-Si:H mixed state. These films typically show relatively high dark
currents at room temperature, which in turn limited the maximum electric BIAS we could apply
to rather low values. Accordingly, the accuracy of the photomixing lifetime and mobility
determinations suffered from both low electrical fields, i.e. low photomixing signals, and, more
seriously, a heating of the sample during measurements that led to an increase of the current of
one order of magnitude over the actual photocurrent.

In order to meet the requirements of high accuracy and stable sample temperature, we modified
our Photoconductive Frequency Mixing setup and are now able to apply BIAS pulses of arbitrary
width and frequency. The dc-current is now monitored sampling the voltage over a 1kΩ resistor
by a PC-DAQ board. The whole setup is controlled using LabVIEW programs.

                                                                                     5500
                    5900

                    5800                                                             5400

                    5700
     Current (µA)




                                                                      Current (µA)




                                                                                     5300

                    5600
                                                                                     5200
                    5500
                                                  (a)                                5100
                                                                                                                       (b)
                    5400

                    5300                                                             5000
                        0   10   20      30   40        50   60                             0   10   20      30       40     50   60

                                      Time (ms)                                                           Time (ms)


 Figure 1. Current through the µc-Si sample MVS723 (supplied by MVSystems, see previous
   section) during one 50ms BIAS pulse under illumination (a) and in dark condition (b).


Figure 1 shows the time dependent dc-current under BIAS application in the dark and
illumination case. The sampling rate here is 1000 samples per second (applies for both dc-current
and ac-signal), one 50ms - pulse is applied per second whereby one measurement cycle consists
of one dark and one light signal acquisition. When the light current is to be measured during a
field dependent measurement, the shutter is opened 100ms in advance, which means a total
illumination duration of 0.25 seconds per two-second-cycle. In the case of decay measurements,
the sample is illuminated 1.75 seconds per cycle. These results are then averaged over at least
100 points for each BIAS point in the field-dependent dc- and ac-curves (see Fig. 2).




                                                                  4
           40                                                    350



           30                                                    300



                                                                 250
           20
Pm ixing




                                                              I ph
                                                                 200
           10


                                                                 150
            0

                                                                 100

           -10
                 40       60       80        100   120               600   800   1000   1200   1400   1600   1800   2000   2200

                               Voltage (V)                                              Field (V cm)
                                                                                                        -1




                      Figure 2. Photomixing and dc – signal averaging and fitting (solid curves)
                                       (sample: MVS723, light-soaked state).


As to be seen in the above figure, given enough data points to average, we obtain curves with
satisfying accuracy whereas the effective power dissipated within the sample is reduced by a
factor of 20 (50ms BIAS vs. 1s per half-cycle). Therefore, electrical fields of much higher
magnitude are possible. Also, as shown in Fig. 1, the slope in the current due to sample heating
during one pulse is well defined so that a linear fit yields the currents at room temperature; in
addition, they yield a good measure for the temperature stability, i.e. a preferred BIAS-width /
frequency ratio.

However, the total illumination duration of about 15 minutes during one field-dependent
measurement makes degradation corrections necessary, i.e. data points have to be corrected
according to the progress of the transport decay at the time they were taken. For this purpose,
decay data that was taken separately is used.




                                                          5
2. Initial experiments on single film samples with coplanar and perpendicular
   contact geometry



                Light beam                         Sample configuration
                                                   for photomixing measurements
                                                   in both in-plane and perpendicular
                               Al contacts
                                                   electric field direction




                 glass substrate                                            TCO layer

                                      a-Si layer            Light beam


            Figure 3. Sample schematics for measurements in two field directions.



It is clear that the electronic transport properties of plain films do not directly scale with the
actual performance of solar cells built with those films as i-layers. Many complications such as
non-uniform electric field, interface effects, etc. result in a dc-photocurrent which is highly
convoluted and therefore no measure for the i-layer properties alone. The i-layer transport
dependency of the photomixing signal, in particular, is obscured by the non-uniform electric
field profile and the contact geometry related capacitance, which acts as a parallel complex
resistance muting the mixing signal. In an effort to separate these contributions we have initiated
first measurements on single a-Si:H films with both coplanar and sandwich contact
configuration.

As described in Fig. 3, Brent Nelson (NREL) has prepared a sample for measurements in both
cross-layer and in-plane electric field direction so that we have the chance to study the impact of
the changed contact geometry on the results of our photomixing measurements. Two substrates,
one plain glass and one TCO-coated, were evaporated with an intrinsic a-Si:H layer in the same
run. The layer thickness is 0.524 microns (L148). Coplanar Al contacts were evaporated in
coplanar configuration (glass substrate) and sandwich configuration (TCO-coated glass
substrate).

Figure 4 shows our preliminary results for both the ac and dc signals. The sample shows
pronounced non-ohmic behavior, which can be due to the TCO—a-Si interface, the aluminum
contacts or both.


                                                      6
                                                                   6                                                 10                                                            3
                               20




                                                                                              'Dark' Current (µA)
                                                                   4                                                                                                               2
         'Dark' Current (µA)




                                                                                                                            5
                               10




                                                                                                                                                                                        Illum. Current (mA)
                                                                   2                                                                                                               1


                               0                                   0                                                        0                                                      0


                                                                   -2                                                                                                              -1




                                                                        Illum. Current (mA)
                         -10                                                                                            -5
                                                                   -4                                                                                                              -2

                         -20                                                                                   -10
                                                                   -6                                                               -0.50     -0.25      0.00       0.25   0.50
                                                                                                                                                      BIAS (V)
                         -30                                       -8                                                                Sample L148
                                                                   -10                                                               (Perpendicular contacts)
                                                                                                                                     Above:
                                        -2       0          2                                                                        'Zoom' measurement
                                                                                                                                     around 0V
                                              BIAS (V)
                                                                                                                            20
                         1600
                                                                                                   Photomixing Power (fW)



                         1400                                                                                               15
Photomixing Power (fW)




                         1200
                                                                                                                            10
                         1000
                               800                                                                                              5

                               600
                                                                                                                                0
                               400
                                                                                                                                      -1.00      -0.75      -0.50      -0.25      0.00
                               200                                                                                                                     BIAS (V)
                                                                                                                 Sample L148 (Perpendicular contacts)
                                    0
                                                                                                                 Photomixing Signal
                                                                                                               Above:
                           -200
                                         -2       0          2                                                 'Zoom' measurement
                                                                                                               around 0V with averaging (solid points)
                                               BIAS (V)

                          Figure 4. DC and AC currents with perpendicular contact geometry(Film thickness 5244Å)



             While, according to B. Nelson and Qi Wang (NREL), it is very difficult to obtain ohmic contacts
             between TCO and intrinsic silicon layers, it seems possible to reduce the accompanying effects
             (field profile distortion etc.) by increasing the layer thickness (which of course is limited by the
             light absorption profile to around 2µm) and choosing contact materials according to their work
             function.




                                                                    7
                                                                                    However, while the dc-curves suggest at least
                                                                                    two non-ohmic contributions, we only find one
                                                                          1.5
                    15
                                                 coplanar contacts with
                                                                                    distinct minimum of the mixing curve, which
                             TCO - a-Si:H - Al
                                                 capacitance correction             indicates a zero-field condition within the
                             configuration
Mixing Power (fW)




                                                                                    intrinsic layer at an external BIAS of –0.5V.
                    10                                                    1.0

                                                                                    Figure 5 shows a comparison between the
                                                                                    mixing power obtained from the perpendicular
                     5                                                    0.5
                                                                                    contact configuration (solid points) and the
                                                                                    sample with coplanar contacts.
                     0                                                    0.0
                                                     For the sake of comparability, we estimated
                         0         5000      10000   15000      20000
                                                     the power losses due to the inherent
                   Electric Field (V/cm)
                                                     capacitance of the sample as 1.2nF (contact
                                                     diameter = 2.5mm, film thickness = 0.54µm,
      Figure 5. Field dependent mixing power
                                                    dielectric constant for a-Si (GHz region) =
      for both configurations. For the TCO–a-
                                                    13.7). Using these assumptions the sample
      Si:H–Al sample, the zero-field point was
                                                    capacitance is 1.2 nF and accordingly, at a
        shifted according to the mixing power
                                                    frequency of 252MHz, the complex resistance
            minimum at V = -0.5V (Fig. 4)
                                                    parallel to the photocurrent source is about 100
   times lower than the setup impedance. With according corrections applied, it turns out that the
   mixing signal of the sample with coplanar contacts is much lower than in the transverse electric
   field case. Moreover, in the case of transverse field direction, we find a rather parabolic field
   dependence of the mixing power (Best seen in Fig. 4, small graph). Both indicate that with the
   electric field in transverse direction most charge carriers contributing to the current are swept
   out. Measurements on different samples with the above mentioned improvements will be carried
   out in order to show whether there is any significant impact of the i-layer preparation on the
   photomixing results.




                                                                                8
3. Hydrostatic pressure dependence of the charge transport in amorphous
   silicon


Models to explain the Staebler-Wronski effect in hydrogenated amorphous silicon have been
proposed which involve the motion of H atoms. Hydrogen motion in turn may be affected by
atomic distances; so it was of interest to study the hydrostatic pressure dependence of the charge
transport parameters in amorphous silicon. By employing the photomixing technique, a
measurement of the hydrostatic pressure dependence of the mobility and lifetime in the annealed
and the light–soaked states could give insight into the dynamics of the Staebler-Wronski effect.
Hydrostatic pressure as well as uniaxial stress and substrate misfit stress measurements has been
performed over the years; however, at first sight many results seem inconsistent or even
contradictory. Many of the measurements employed diamond anvil techniques, which result in
uniaxial stress that varies bond angles rather than bond distances; one would expect that the
mobility of the hydrogen atoms would be affected by the change in the lattice distance.

For our photomixing measurements we employed a clamped pressure cell with a quartz window.
The pressure transmitting fluid was 3M Fluorinert, an electronic fluid with flat optical
transmission throughout the visible region and high dielectric strength. The latter is of
importance since it is necessary to apply high electric field to the samples in order to measure the
electric field dependence of the mobility and so determine the range and the depth of the long-
range potential fluctuations. For these initial measurements, an a-Si:H sample produced by the
hot-wire technique with 7-9% H was employed.

Figure 6 shows the results of a first pressure dependent photoconductivity measurement of a-
Si:H during light soaking. The starting points show that the photocurrent drops with increasing
pressure. But also the decay is reduced under higher pressure. The 3 kBar and the 1 kBar curves
cross at about 100 minutes which means that after that time the photoconductivity, though
starting at a lower value, is still higher after long time irradiation. Figure 7a shows the pressure
dependence of the drift mobility as a function of illumination time indicating that at 3 kBar it
becomes independent of illumination time. Figure 7b shows the pressure dependence of the
lifetime as function of illumination. The lifetime curves though also starting at lower values
under pressure show pressure independent slopes.

From this we can conclude that the generation rate of recombination centers during light
exposition is pressure independent, as well. However, the number of dangling bonds seems to
rise under pressure (starting points), which supports the collapsing-void-presumption. The
different decay rates of mobility and lifetime under pressure indicate that the rate of generation
of charged and neutral scattering defects vary with pressure.

Summarizing our first pressure results one can think of two different ways to describe the
pressure dependency of the SW-effect:




                                                 9
                                                                   12



                                    Photoconductivity (10 Ω cm )
                                    -1
                                                                            0 Kbar
                                                                   11
                                                                                                                                   annealing

                                    -1
                                                                   10                                                              again
                                    -4                              9       1 Kbar
                                                                    8
                                                                    7
                                                                    6        3 Kbar
                                                                    5
                                                                    4
                                                                    3
                                                                                 0.1          1                        10           100
                                                                                       Illumination time (min)
                                Figure 6. The Photoconductivity vs. Illumination time for three different pressures.

                The temporal order of the measurements: 0 kBar ⇒ 1 kBar ⇒ 3 kBar ⇒ 0 kBar (“annealing
                again”). After each turn the sample was reannealed under atmosphere pressure for 1 h at 150 0C
                and then put into the pressure cell again.

                                                                                                                  400
                         0.36                                                                                             0 Kbar
                                                                                  (a)                             350
                         0.32
Drift mobility (V/cm )
2




                                                                                                                        1 Kbar
                                                                                                                  300
                                                                                                  Lifetime (ns)




                         0.28

                         0.24                                                                                     250   3 Kbar
                                                                   0 kBar
                         0.20
                                                                   1 kBar                                         200
                         0.16                                      3 kBar                                                           (b)
                                                                                                                  150
                         0.12
                                      1                                     10          100                        0.01       0.1         1      10        100
                                                        Illumination time (min)                                                  Illumination time (min)

                                                                        Figure 7. The drift mobility and lifetime as functions
                                                                           of illumination time under different pressures.

                Under pressure voids can collapse and weak bonds can break resulting in additional dangling
                bonds reducing both the dark current and photoresponse. This would mean that pressure does
                more or less the same as light illumination does: it creates local defects. The pressure
                dependence of the light soaking experiments may then be regarded as time–shifted. That means,
                even before illumination there are as many pressure induced defects as though the sample would
                have been exposed to light as long as it takes to turn down the photoresponse to the value under


                                                                                                                  10
pressure at the start. These additional dangling bonds result in a decrease in mobility and
consequently a decrease in the photoresponse.

It seems possible that pressure application first introduces local defects that can be observed
promptly in a change of lifetime and mobility; additionally, pressure related lattice variations
change the time dependence of the mobility during illumination, which may indicate different
generation rates of charged defects. We plan to perform further pressure experiments on samples
prepared by different preparation techniques to resolve these questions.



Change of the film thickness after pressure application

Further measurements on the pressure dependence of the transport parameters of amorphous
silicon films were performed, with emphasis on possible defects introduced by pressure
application. We show both FTIR- and photomixing measurements on amorphous silicon
samples.

We measured the change of the IR - spectra of the sample T516 (Qi Wang) first in the annealed
state, then after the application of hydrostatic pressure of 3 kBar. After this, we annealed the
sample for one hour and repeated the FTIR measurement. From the interference fringes
occurring in the spectra it is basically possible to determine the film thickness. The fringes occur
as a result of a small refraction index mismatch at the film–substrate-interface (n=3.0 ↔ n=3.4).
From the periodicity one can easily determine the respective film thicknesses. Figures 8 and 9
show both the fringes from which the thickness was determined and the results for the respective
cases. The initial value of 4.408 µm is in good agreement with the value of 44,100Å given by Qi
Wang. After pressurizing the sample the film seems to remain around one hundred nanometers
thinner than in the initial state, which would mean a decline of 2.3%. This number seems quite
high, though; on the other hand, the apparent thickness change may also be due to a change of
the index of refraction. Though reannealing for one hour recovers the sample in part, the
thickness remains 40 nm under the initial thickness. However, the results show permanent
structural changes introduced by pressure application.

From the actual hydrogen related absorption peaks we could not detect any change in the
hydrogen configuration, at least as far as the 2000/2100 – double peak is concerned.




                                                11
                                                                                                                                                                                 4.42

                                                                                 ann                                                                                             4.40
                                                    100
                                                                               press                                                                                             4.38




                                                                                                                                                                Thickness (υm)
                                                                              reann
                                Transmission (%)
                                                                                                                                                                                 4.36

                                                                                                                                                                                 4.34

                                                                                                                                                                                 4.32

                                                                                                                                                                                 4.30
                                                                                                                                                                                                               R
                                                                                                                                                                                    Annealed Pressurized (3kBar) eannealed
                                                                                                                                                                                                        State
                                                    90

                                                                            Wavenumber                                                                                            Figure 9. T516 – film thickness
                                                                           2000                      3000                                                                           after pressure application.
                                                                                  Waveno
                                  Figure 8. Infrared transmission of T516
                                under annealed (“ann”), pressured (“press”),
                                   and reannealed (“reann”) conditions.

                                                                                                                                                      500
                                22
 Photoconductivity (10 /Ω cm)




                                                                            0.2                                                                                                  0 kBar
                                20                                                  0.65
                                                   0 kBar                                                                                             400
                                                                                                                            Photomixing Signal (fW)
                                                                                           0.16
                                18                                                                 0 kBar
-4




                                16                                                                                                                                                                                          0 kBar
                                                               2.63          2.71
                                                                                                                                                      300
                                                                                                                                                                                                       0.2
                                14                                                   2.71                                                                                                                      0.65 0.16
                                12                                                           2.8                                                      200
                                10
                                   8                                                                                                                  100
                                                                                                                                                                                         2.63
                                   6                                                                                                                                                                     2.71
                                                                                                                                                                                                                 2.71
                                                                                                                                                                                                                           2.8
                                   4                                                                                                                     0
                                   0.01                              0.1                      1                                                          0.01                                    0.1                        1
                                                              Illumination Time (min)                                                                                                     Illumination (min)
                                1.2                                                                                                                   650
                                1.1                                                                                                                   600
                                1.0                                                                                                                   550                                                             2.8
                                0.9                                                                                                                                                                            2.71
                                                                                                                                                      500                                               2.71
Mobility (cm /Vs)




                                                            0 kBar                            0 kBar                                                                                    2.63
                                0.8
                                                                                                            Lifetime (ns)




                                                                                                                                                      450
2




                                0.7                                          0.2 0.65
                                                                                           0.16                                                       400
                                0.6
                                0.5
                                                                                                                                                      350                                              0.2 0.65 0.16
                                                                                                    0.44                                                                                                             0 kBar
                                0.4                           2.63                                                                                    300      0 kBar
                                                                              2.71 2.71
                                0.3                                                                                                                   250
                                                                                            2.8
                                0.2                                                                                                                   200
                                0.1                                                                                                                   150
                                  0.01                               0.1                      1                                                         0.01                                   0.1                     1
                                                              Illumination Time (min)                                                                                                    Illumination (min)

       Figure 10. Pressure-dependent transport parameters. After initial annealing, the sample was
       pressurized and released again. We performed four cycles in the same way, with the transport
       parameters monitored in situ. The x-axis position is associated with the over-all illumination
                                     period during the measurement.




                                                                                                            12
                         Transport vs. Pressure Hysteresis Measurements

                         Four cycles with a pressure from 0 to 2.8 kBar being applied and released were run during the
                         measurement. The photoconductivity, mobility, and lifetime were measured at each stage. The
                         results are shown in the Figure 10. It was found that under a given pressure the
                         photoconductivity and mobility decreased while the lifetime increased. However, when the
                         pressure was released a partial recovery of the photoconductivity and mobility was observed but
                         not to the values obtained at the initial zero pressure. These results indicate that there are two
                         effects of hydrostatic pressure; one an elastic effect where the mobility decreases under pressure
                         but is restorable when the pressure is released and an inelastic effect where the mobility is
                         decreased due to the introduction of charge scattering centers.

                         In order to illucidate the participating effects in amorphous silicon under pressure we also
                         performed a similar measurement on single crystalline silicon with the results shown in Figure
                         11. They show that for single crystals there is only an elastic effect.


                                                                                             300
                                                                                             280
                                                                                             260       1.1kBar   0.36kBar
                       220                                                                   240
DC photocurrent (µA)




                                                                        Photomixing signal




                                                                                             220
                                                                                             200                               0kBar
                                                                                             180
                                                                                             160
                                                                                                           3.05kBar 2.91kBar
                                                                                             140

                       200                                                                   120
                                                                                             100
                                                                                              80
                                                                                              60
                                                                                              40
                                                                                              20
                                                                                               0
                       180                                                                         0       20    40     60     80      100
                             0      20     40     60     80     100

                                 Number of measurement points                                      Number of measurement points


                                     Figure 11. DC photocurrent and photomixing signal under different pressures
                                          in single crystalline silicon. Both decrease with increasing pressure.
                                                         The curves show only an elastic effect.




                                                                           13
4. Hydrostatic Pressure Dependence of Charged Transport and Small Angle
   X-ray Scattering Measurements



We had previously reported on a series of measurements on the charge transport properties of
amorphous silicon determined by photomixing as a function of hydrostatic pressure. It was found
that under a given pressure the photoconductivity and mobility decreased while the lifetime
increased. However, when the pressure was released a partial recovery of the photoconductivity
and mobility was observed but not to the values obtained at the initial zero pressure. These
results indicate there are two effects of hydrostatic pressure; one an elastic effect where the
mobility decreases under pressure but is restorable when the pressure is released and an inelastic
effect where the mobility stays decreased due to the introduction of charge scattering centers. It
was conjectured that a permanent collapse of the voids in a-Si:H may be responsible for the
inelastic component.

To this end samples were supplied by Don Williams, which we subjected to hydrostatic pressure.
Subsequently, the SAXS measurements were performed by Don Williams at the Colorado
School of Mines. The samples consisted of one low temperature preparation with a relatively
high void fraction (about 2 vol%) and one a high substrate temperature with a very low void
fraction (≤ 0.02 vol%). The SAXS data showed no evidence for residual microvoid collapse.
Therefore, the origin of the inelastic component is still unknown. Other changes to be considered
are bond angles and bond lengths.




                                               14
5. The “uninterrupted growth/annealing” method

During the last period a series of a-Si single film samples supplied by Guanglin Kong1 was
measured by means of the Photoconductive Frequency Mixing method. The series consists of
four samples with varied deposition temperature (G179 – G204), two samples which were
prepared for the investigation of light-induced structural changes, and two more samples
prepared by means of the “uninterrupted growth/annealing” PECVD method, which was
developed by the Kong-group.

In contrast to the “Chemical Annealing” or layer-by-layer deposition where growth and H+
annealing alternate, the hydrogen treatment happens during the deposition process. This
approach which essentially amounts to hydrogen dilution as has been studied for years by, e.g.,
Tsu [7]. As also suggested by Tsu, the hydrogen plasma is kept at a level near the threshold to
the formation of microcrystalline silicon. At this level, the etching effect of hydrogen during the
growth process results in a more dense and stable Si network without the formation of larger
microcrystalline regions. This leads to amorphous films with improved stability yet without the
typical drawbacks of mixed or µc-Si films, i.e. lower optical response due to partly indirect
transition, and lower mobility due to grain boundaries.

A slight boron compensation is believed to shift the Fermi level down towards the midgap which
results in a higher photosensitivity as well as a higher electron mobility. CPM measurements on
appropriately compensated a-Si samples also suggest a lowered subgap absorption which might
be due to a passivation of native donor-like impurities through the formation donor-acceptor
pairs, compensation of dangling bonds and release of local strain in the Si network.

According to [8] and [9],appropriate hydrogen dilution results in amorphous silicon films with
light-induced decay of the photocurrent of about one half order of magnitude with saturation
tendency after 104 seconds of illumination. With additional boron compensation, the samples
show essentially stable transport properties. In comparison, an a-Si:H standard sample showed a
photocurrent decay of about one order of magnitude. With samples where both hydrogen dilution
and boron compensation were applied, even an increase of the photocurrent was found in some
samples [8] which is explained in terms of an additional light-induced effect which involves a
bond-switching from fourfold to threefold B configurations forming neutral acceptor dangling
bond pairs. These, in turn, lower the B doping efficiency and therefore, in the case of slight B
compensation, result in a shift of Ef towards Ec and accordingly an increase of the electron
mobility.




1
    Institute of Semiconductors, Chinese Academy of Sciences, Beijing, China


                                                         15
Photomixing Measurements on samples supplied by Guanglin Kong

The sample properties are given in Table 1.

Table 1.      Deposition conditions

 Sample          SiH4:H2   Ingredient         Tsubstr.   Tdep    Thickness    Deposition
 ID                                           (oC)       (hrs)   (µm)         Rate (Å/s)
 I 70426         1:1                          360        4.5     1.5          0.93
 I 80929         1:3                          300        3.5     3            2.40
 P 60715 I       1:12      0.2% B2H6(10-4)    225        3.0     0.89         0.82
 P 60717 II      1:12      2% B2H6(10-4)      225        2.6     0.79         0.84
 G 179           1:5                          350        1.5     1.25         2.31
 G 202           1:5                          400        4.0     1.7          1.18
 G 203           1:5                          300        5.0     3.2          1.78
 G204            1:5                          200        3.0     2.2          2.04


As shown in Table 1, a series of four samples, G179 - G204, is prepared with the substrate
temperature being varied at constant dilution ratio. The respective light-induced decay curves are
shown in figures 12 and 13 beginning with the highest substrate temperature. For all samples, the
mobility after three hours of illumination lies somewhere between 50% and 60% of the initial
value. The lifetime decay, however, seems to scale with the deposition temperature, i.e. the lower
the deposition temperature the lower the final lifetime with respect to the initial value. Also the
initial photomixing mobility is clearly higher for films deposited with higher Ts. For the
photoconductivity this means that for lower deposition temperatures (Ts < 350oC) both the initial
value (due to the lower initial mobility) and the decay rate (due to the higher decay of the
lifetime) turn out to be rather poor as compared to samples with Ts ≥ 350oC. It seems that given a
moderate dilution ratio the network reconstruction during growth requires more activation
through the substrate temperature than, e.g. in the case of Hot-Wire CVD. From this series it is
clear that the best sample is also the most expensive one in terms of substrate temperature and
deposition rate.




                                                16
                                                                                                      0.25

                            1500
                                                                                                      0.20
                            1250




                                                                                Mobility (cm / V s)
                                                                                                      0.15
Lifetime (ns)
                            1000




                                                                              2
                                        750
                                                                                                      0.10
                                        500
                                                                                                      0.05
                                        250

                                             0                                                        0.00
                                                 0.1     1      10   100                                      0.1     1      10    100
                                                        Time (min)                                                   Time (min)

                                            40
            Photoconductivity (10 / Ω cm)




                                            30

                                                                                                              Sample G202
        -4




                                            20

                                                                                                         Tdepos. = 400 oC
                                            10



                                             0
                                                 0.1     1      10   100
                                                        Time (min)


                                                                                                      0.50
                                    900
                                                                                                      0.45
                                    800
                                                                                                      0.40
                                    700                                                               0.35
                                                                              Mobility (cm / V s)




                                    600                                                               0.30
   Lifetime (ns)




                                                                              2




                                    500                                                               0.25
                                    400                                                               0.20
                                    300                                                               0.15
                                    200                                                               0.10
                                    100                                                               0.05
                                             0                                                        0.00
                                                  0.1     1     10   100                                       0.1     1      10   100
                                                        Time (min)                                                   Time (min)



                                            60
        Photoconductivity (10 / Ω cm)




                                            50                                                                 Sample G179
      -4




                                            40

                                            30
                                                                                                             Tdepos. = 350 oC
                                            20

                                            10

                                            0
                                                  0.1     1     10   100
                                                        Time (min)
                                            Figure 12. Samples G202 and G179, dilution ratio is kept constant at 1:5



                                                                               17
                   1200
                                                                                                                   0.12

                   1000                                                                                            0.10




                                                                                        Mobility (cm / V s)
                               800                                                                                 0.08
Lifetime (ns)




                                                                                        2
                               600                                                                                 0.06

                               400                                                                                 0.04

                               200                                                                                 0.02

                                                  0                                                                0.00
                                                          0.1       1       10    100                                        0.1      1       10    100
                                                                                                                                     Time (min)
                                                                        Time (min)
                                                  7
                Photoconductivity (10 / Ω cm)




                                                  6

                                                  5
                                                                                                                          Sample G203
                -4




                                                  4
                                                                                                                     Tdepos. = 300 oC
                                                  3

                                                  2

                                                  1

                                                  0
                                                          0.1       1       10    100
                                                                   Time (min)


                                        1000                                                                       0.045
                                                                                                                   0.040
                                                  800                                                              0.035
                                                                                             Mobility (cm / V s)




                                                                                                                   0.030
         Lifetime (ns)




                                                  600
                                                                                                                   0.025
                                                                                          2




                                                                                                                   0.020
                                                  400
                                                                                                                   0.015
                                                                                                                   0.010
                                                  200
                                                                                                                   0.005

                                                      0                                                            0.000
                                                            0.1         1        10                                            0.1        1        10

                                                                   Time (min)                                                         Time (min)


                                                  3.0
                  Photoconductivity (10 / Ω cm)




                                                  2.5                                                                          Sample G204
                -4




                                                  2.0
                                                                                                                           Tdepos. = 200 oC
                                                  1.5

                                                  1.0

                                                  0.5

                                                  0.0
                                                            0.1         1        10
                                                                  Time (min)
                                                Figure 13. Samples G203 and G204, dilution ratio is kept constant at 1:5


                                                                                         18
                                                                                                                  0.4
                          1000

                                                                                                                  0.3




                                                                                         Mobility (cm / V s)
                                       800
Lifetime (ns)




                                                                                         2
                                       600
                                                                                                                  0.2

                                       400
                                                                                                                  0.1
                                       200

                                                 0                                                                0.0
                                                     0.1    1       10    100                                           0.1   1      10    100


                                                           Time (min)                                                         Time (min)

                                                60


                                                                                                                  Sample I70426
          Photoconductivity (10 / Ω cm)




                                                50

                                                40
        -4




                                                                                            Tdepos. = 360 oC
                                                30
                                                                                            SiH4 : H2 = 1:1
                                                20

                                                10

                                                0
                                                     0.1    1       10    100

                                                           Time (min)


                               1200
                                                                                                                  0.4
                               1000
                                                                                            Mobility (cm / V s)




                                                                                                                  0.3
  Lifetime (ns)




                                            800
                                                                                         2




                                            600                                                                   0.2

                                            400
                                                                                                                  0.1
                                            200

                                                 0                                                                0.0
                                                     0.1        1    10     100                                         0.1   1      10    100

                                                           Time (min)                                                         Time (min)

                                                40
                Photoconductivity (10 / Ω cm)




                                                30
                                                                                                                   Sample I80929
           -4




                                                20
                                                                                                  Tdepos. = 300 oC
                                                                                                  SiH4 : H2 = 1:3
                                                10



                                                 0
                                                     0.1     1       10     100
                                                           Time (min)

                                                                          Figure 14. Samples I70426 and I80929.



                                                                                                               19
                                                  400
                                                                                                                                0.25

                                                  300                                                                           0.20




                                                                                                          Mobility (cm / V s)
              Lifetime (ns)



                                                                                                                                0.15




                                                                                                        2
                                                  200

                                                                                                                                0.10

                                                  100
                                                                                                                                0.05


                                                       0                                                                        0.00
                                                           0.1     1     10    100                                                     0.1       1      10    100
                                                                  Time (min)                                                                     Time (min)

                                                      20
                                                                                                                                 Sample P60715I
                      Photoconductivity (10 / Ω cm)




                                                                                                                            Tdepos. = 225 oC
                                                      15
                   -4




                                                                                                                            SiH4 : H2 = 1:12
                                                      10
                                                                                                                            +0.2% B2H6(10-4)
                                                       5



                                                       0          Time (min)
                                                           0.1     1     10    100
                                                                                                                   0.125
                                     140                                             Time (min)
                                     120                                                                           0.100
                                                                                        Mobility (cm / V s)




                                     100
Lifetime (ns)




                                                                                                                   0.075
                                                                                        2




                                                      80
                                                      60                                                           0.050
                                                      40
                                                                                                                   0.025
                                                      20
                                                       0                                                           0.000
                                                            0.1     1     10   100                                      0.1                  1         10     100
                                                                  Time (min)                                                                     Time (min)
  Normalized Photoconductivity




                                            1.0

                                            0.8                                                                                    Sample P60715I
                                                                                                                                 (rough surface)
                                            0.6

                                            0.4
                                                                                                                                 Tdepos. = 225 oC
                                                                                                                                 SiH4 : H2 = 1:12
                                            0.2                                                                                  +0.2% B2H6(10-4)
                                            0.0
                                                            0.1    1      10   100
                                                                  Time (min)

                Figure 15. Samples P60715I with flat and rough surface.
   These are prepared by employing the “uninterrupted growing/annealing” technique.



                                                                                                                            20
From Figure 14, we can see that samples I70426 and I80929 again suggest that the substrate
temperature has a large impact on the film stability. The sample with lower hydrogen dilution
ratio but higher substrate temperature shows better stability at a higher performance level. The
deposition rate of I80929, on the other hand, is about three times that of I70426. This means that
other parameters not specified by the Kong group may have been changed, too. However, the
thicknesses of I70426 and I80929 were just estimated from the deposition period.

Finally, Figure 15 shows our results for the two samples prepared by “uninterrupted
growth/annealing” PECVD, P60715I and P60717II. Actually, we had two ‘P60715I’ samples
available, one with a flat and one textured surface. Both P60715I samples show a decay of all
transport quantities in the range of 20% (mobility) through about 50% - 60% (lifetime,
photoconductivity). However, a first slight increase in mobility is observable (P60715I – flat
surface) but after about 15 min the mobility begins to decrease, too. In comparison to the
samples with mere dilution and from our results of the light-induced decay, we cannot detect any
noticeable impact of boron doping on transport for the present samples other than what is
mentioned above, but samples are more stable.

Unfortunately, we were unable to detect any reasonable mixing signal on the P60717II. This is
particularly disappointing since this would have been the only sample that we have access to
independent data for [8].

Table 2 below summarizes the Long-range potential fluctuation data we acquired for the whole
series. The values for LRPF range and depth reflect the good stability of the mobility in most
samples, particularly I70426 and I80929.


Table 2.      Long range potential fluctuations

                            Substrate
 Sample ID     H-Dilution                   Range (nm)       Range (nm)        Depth (eV)
                            Temperature
                                            Annealed         83.3               0.10
 I 70426       1:1          360
                                            Light-soaked     83.3               0.11
                                            Annealed         89.0               0.10
 I80929        1:3          300
                                            Light-soaked     88.7               0.11
               1:12                         Annealed         30.1               0.11
 P60715I                    225
               +B2H6                        Light-soaked     30.1               0.12
 P60715I       1:12                         Annealed         39.0               0.13
                            225
 (rough s.)    +B2H6                        Light-soaked     38.0               0.13
                                            Annealed         19.1               0.13
 G179          1:5          350
                                            Light-soaked     16.3               0.14
                                            Annealed         40.8               0.09
 G202          1:5          400
                                            Light-soaked     46.4               0.13
                                            Annealed         24.2               0.12
 G203          1:5          300
                                            Light-soaked     25.8               0.15
                                            Annealed         Weak mixing signal /
 G204          1:5          200
                                            Light-soaked     Low mobility




                                                  21
6. Charge transport properties near and above the transition from
   amorphous to microcrystalline silicon


6.1   Material produced at NREL

Microcrystalline silicon (µc–Si:H) has been the object of studies for many years to understand its
electronic, optical and structural properties. However, recently there has been a renewed interest
in incorporating µc – Si:H in different solar cell structures since such devices have shown no
degradation after prolonged light soaking [7,10]. It has been shown that a-Si:H films grown
under hydrogen dilution close to the onset of microcrystallinity exhibit a higher degree of
stability against light soaking compared to a-Si:H. In hot-wire assisted chemical vapor deposition
(HW-CVD) [11-15], the decomposition of silane and hydrogen gas mixture allows one to
prepare material with a transition from to amorphous to microcrystalline growth by variation of
the silane to hydrogen dilution. By varying the grain size, an enhanced absorption of
microcrystalline compared to crystalline silicon has been observed [16]. The (HW-CVD) method
allows a continuous preparation of material with a smooth transition from amorphous
hydrogenated silicon to microcrystalline silicon. The photoconductivity in microcrystalline
silicon has been studied as a function of Fermi-level to determine the mobility-lifetime product
(µτ) [17] with the observation that (µτ) increased significantly by shifting the Fermi level from
the mid-gap towards the conduction or valence band; this was attributed to the increase in the
majority carrier lifetime due to a change in the thermal occupation of defect centers by the shift
of the Fermi level. It is of interest to separately determine the mobility and lifetime in the
transition from the amorphous to the microcrystalline state; dc photoconductivity measurements
determine the µτ product. The independent determination of µ and τ was accomplished by
employing the photoconductive frequency mixing technique [1-6]. By observing the increase in
the drift mobility as a function of electric field, the range and the depth of the long-range
potential fluctuations [4] as a function of hydrogen content and hence the change in these
quantities in the transition from amorphous to microcrystalline states were determined.

In this report we present results from electronic transport measurements of a series of samples
prepared employing the hot-wire chemical vapor deposition technique supplied by Qi Wang.
Within this series a transition from the amorphous to the microcrystalline state is realized. Since
there is evidence that one finds improved stability against light-induced degradation in the mixed
state it is interesting to study the (photo-) transport properties and stability with respect to this
transition. Table 3 shows a summary of the sample characterization:




                                                 22
Table 3.   Summary of the sample characterization results (Qi Wang).

   Sample H/SiH4          XRD          (µc) av.      Sample      H-content Deposition
   ID                     Structure    Grain         Thickness   (%)       rate (Å/s)
                                       size (nm)     (Å)
   T516      1            a-Si                       41,400      13.3          17.25
   T517      5            µc Si        12            21,350      5.2           5.93
   T518      10           µc Si        13            14,300      4.7           3.97
   T519      20           µc Si        14            11,200      3.9           2.33
   T528      5            µc Si        18            25,400      3.1           7.05
   T529      4            µc Si        12            24,000      4.2           8.00
   T530      3            a-Si+µc Si   9             16,400      4.4           7.80
   T531      2            a-Si                       25,000      8.9           10.41
   T532      3            a-Si+µc Si   11            17,000      4.0           7.08
   T534      2            a-Si                       30,000      10.9          12.50


Results of photomixing measurements

We employed the photoconductive frequency mixing technique to determine the transport
properties in terms of the charge carrier (photomixing) lifetime and mobility as well as the over-
all photoresponse.

In a-Si:H, the electronic transport depends tightly on the hydrogen content, so the
photoconductivity is usually plotted against the amount of H-atoms introduced into the samples.
A look at Table 3 indicates that within this series of samples microcrystalline structure results in
a low hydrogen content. Therefore, we decided to plot the transport properties against the
hydrogen content, including all given samples, as shown in Fig. 16.

With the decrease of the H-content the photoconductivity is more or less constant as long as the
hydrogen concentration is over about 5% to 6%. However, it shows a sudden increase when the
H-concentration drops below 5%, i.e. the microcrystalline regime is entered.




                                                23
                                          400

           Photoconductivity (10 Ω cm )
           -1
           -1
                                          300
           -4




                                          200


                                          100


                                           0

                                                2     4     6     8     10     12   14
                                                             H content (%)

                                                    Figure 16. Photoconductivity of amorphous and
                                                     microcrystalline silicon vs. hydrogen content.




To investigate what quantity this increase is due to, we determined the photomixing lifetime
τ and mobility µ for each sample in the annealed state. The results are shown in Figures 17 and
18. Both τ and µ are affected by the transition. Similarly to the behavior of the
photoconductivity, the lifetime exhibits a sudden increase up to two orders of magnitude when
the hydrogen content drops to values less than 5% (Figure 17). The mobility, however, shows
somewhat opposite behavior. In the amorphous region, the mobility increases with decreasing H-
concentration. This is surprising since one usually finds the mobility to increase when the
hydrogen content grows. As a comparison, Figure 19 shows the results for a set of samples also
produced by hot-wire CVD, but entirely within the amorphous region. For H-contents under 5%,
the mobility drops to values not detectable anymore. However, except for the samples T518 and
T528, µ only drops to about 20% of its maximum value, so the photoconductivity, being straight
proportional to τ and µ, still increases dramatically.




                                                                          24
                14000
                12000
Lifetime (ns)   10000
                        8000
                        6000
                        4000
                        2000
                              0
                   -2000
                                  2       4      6       8        10   12     14
                                                    H content (%)

                              Figure 17. The photomixing lifetime of amorphous and
                                   microcrystalline Si:H vs. hydrogen content.




                            0.5

                            0.4
                                      µc-Si
        Mobility (cm /Vs)




                            0.3
                                      Region
     2




                            0.2
                                                         a-Si:H
                            0.1                          Region
                            0.0

                                  2      4      6       8         10   12    14
                                                  H content (%)

                                       Figure 18. The mobility of amorphous and
                                       microcrystalline Si:H vs. hydrogen content.




                                                             25
                                                                        10-12%
                                                                  3.0                    7-9%




                                       Drift mobility (cm V s )
                                                                                                                       2-3%




                                      -1 -1
                                                                  2.5
                                      2                                                         6-7%

                                                                  2.0                                                      2-3%




                                                                  1.5                                                                 <1%




                                                                  1.0
                                                                     280         300 320 340 360 380 400
                                                                                                                   o
                                                                                 Deposition substrate temperature ( C)

                                   Figure 19. Drift mobility in the case of entirely amorphous samples
                                            (Hot-wire assisted PECVD a-Si:H from NREL).

Long-Range Potential Fluctuation

Long-range Potential Fluctuations (LRPF) are thought to reflect the density and arrangement of
charged defects in amorphous silicon. To determine the LRPF within the samples, we performed
field-dependent measurements of the mobility and lifetime. The results are shown in Figures 20
and 21. Note that the range increases to values of around 50nm in the µc – regime, which might
reflect the occurrence of grain boundaries. Also the depth of the potential fluctuations increases,
whereas both quantities decrease with decreasing hydrogen content for H-concentrations over
6% (i.e. the amorphous regime).

These results lead to a possible explanation for both the drop of the mobility and the dramatic
increase of the lifetime in the case of microcrystalline silicon. Grain boundaries might serve as
scattering centers as well as barriers against recombination.

                      70
                                                                                                                      0.14
                      60
                                                                                                                      0.13
 Range of LRPF (nm)




                                                                                                 Depth of LRPF (eV)




                      50
                                                                                                                      0.12
                      40                                                                                              0.11

                      30                                                                                              0.10

                      20                                                                                              0.09

                           2   4      6                           8      10        12   14                                   2    4    6    8     10  12   14
                                        H content (%)                                                                                   H content (%)

                                      Figures 20 and 21. Range and depth of long-range potential
                                   fluctuations in amorphous and microcrystalline silicon-hydrogen.


                                                                                                                      26
Light-induced Decay Measurements

This section covers measurements of the light-induced degradation (Staebler-Wronski-effect) for
samples in the amorphous state, mixed state (a-Si + µc-Si) and the microcrystalline state (µc-Si).
Figure 22 shows the normalized photoconductivity for the three amorphous samples, which
exhibit behavior as expected from former results. In each case the samples were first annealed
for 1 hour at a temperature of 150 0C.



                                                 1.1
                                                                        T516
                  Normalized photoconductivity




                                                 1.0
                                                                        (13.3%, 30mW)
                                                 0.9

                                                 0.8

                                                 0.7    T531
                                                 0.6    (8.9%, 31.2mW)
                                                 0.5    repeated meas.

                                                 0.4
                                                                    T534
                                                                    (10.9% H, 31.5mW)
                                                 0.3
                                                       0.1          1          10      100
                                                             Illumination time (min)

      Figure 22. Normalized photoconductivity vs. illumination duration for T516, T531
          and T534 (amorphous state). The light intensities are given in the Figure.


Since in the case of microcrystalline silicon we found both high dark and photocurrent, we had to
restrict the number of points in order to prevent those samples from heating. Also, we could not
measure lifetime and mobility at high fields. This results in an increased error and so a certain
scattering of both lifetime and mobility. Nonetheless the Figures 23-25 show increased stability
against light-induced defects for samples in both the mixed and microcrystalline phase. The
increase of the photocurrent in the case of the samples T529 and T532 is probably due to a slight
heating.




                                                                          27
                                     1.4
                                     1.3
                                     1.2


           Normalized decay (a.u.)
                                     1.1
                                     1.0
                                     0.9
                                     0.8
                                     0.7
                                     0.6
                                           QiWang T529 ( µc-Si)
                                     0.5        Photoconductivity
                                     0.4        Drift mobility
                                     0.3        Lifetime
                                     0.2
                                     0.1
                                     0.0
                                                   1                10      100
                                                  Illumination time (min)

Figure 23. Photoconductivity, mobility and lifetime vs. illumination duration, T529.




                                     1.3
                                     1.2
                                     1.1
           Normalized decay (a.u.)




                                     1.0
                                     0.9
                                     0.8
                                     0.7
                                     0.6
                                           QiWang T530 (a-Si + µc-Si)
                                     0.5
                                                Photoconductivity
                                     0.4
                                                Drift mobility
                                     0.3
                                                Lifetime
                                     0.2
                                     0.1
                                     0.0
                                            1                   10            100
                                                  Illumination time (min)

Figure 24. Photoconductivity, mobility and lifetime vs. illumination duration, T530.




                                                                    28
                                   1.4


         Normalized decay (a.u.)   1.2

                                   1.0

                                   0.8

                                   0.6
                                         QiWang T532 (a-Si + µc-Si)
                                   0.4     Photoconductivity
                                           Drift mobility
                                   0.2     Lifetime

                                   0.0
                                              1                10      100
                                             Illumination time (min)

     Figure 25. Photoconductivity, mobility and lifetime vs. illumination duration, T532.



Summarizing the above results, we found both increased stability and a huge increase of the
photomixing lifetime resulting in an increase of the photoresponse for samples entering the
microcrystalline regime. The increase in lifetime and decrease in mobility as the microcrystalline
regime is entered indicates that grain boundaries can serve as scattering centers as well as
barriers against recombination. In order to further illucidate the mechanisms responsible for the
dramatic change in lifetime and mobility as one makes the transition from the amorphous to the
microcrystalline state, it would be useful to perform similar measurements on samples prepared
by different methods which result in different microstructures.




                                                          29
FTIR measurements

                         110


                         100


                         90
          transmission




                         80


                         70


                         60


                         50


                         40
                               0   1000        2000         3000           4000

                                                waveno
                   Figure 26. IR – spectra, sample T516 (a-Si:H) on polycrystalline substrate.

Figure 26 shows a typical a-Si:H spectrum we acquired in the wavelength region from 400 to
4000 cm-1. The total hydrogen content was obtained from the rocking modes of hydrogen in all
possible bonding configurations which give rise to an absorption peak at 640cm-1. To calculate
the hydrogen content from the integrated absorption of this peak we used values for the
oscillation strength parameters given in [18]:

   N H = AI ,
   where N H is the hydrogen                             A640 = (2.1 ± 0.2) * 1019 cm −2
   concentration and                         and         A2000 = (9.0 ± 1.0) * 1019 cm − 2
              α
   I =∫         dω                                       A2100 = (2.2 ± 0.2) * 10 20 cm − 2
              ω
  is the integrated absorbance.

We found good agreement with the values for the hydrogen content given by NREL. The
stretching modes at 2000 and 2100cm-1 are related to isolated SiH in a Si network or SiHx,
hydrogen clusters, and also SiH bondings at surfaces (e.g. voids) whereas only the latter group
gives rise to a 2100 – peak. Figure 27 shows an example of the stretching mode related double-
peak in the case of a-Si:H and Figure 31 contains the respective deconvolution.

In microcrystalline silicon most hydrogen is bonded at grain boundaries facing intergrain voids
and so gives rise to usually at least two peaks around 2100 with respect to the crystalline grain
orientation (figure 28). The intensity ratio of the 2000 and 2100 peaks depends on the dilution as
can be seen in figures 28 and 30.



                                                       30
In the mixed state samples, also bending mode related absorption at around 900cm-1 occurs as
shown in figure 29.

                    900

                    800                                                          100
                                                                                 90       µc Si:H
                    700                              a-Si:H
                                                                                 80       (Sample T519)
                    600                              (Sample T516)               70
                                                                                          H/SiH4 = 20           (220)-, (111)-
                    500                                                          60                             orientation of
                                                                                 50                             crystallites
                    400
   α




                                                                            α
                                                                                 40
                    300                                                          30
                    200                                                          20
                                                                                 10
                    100
                                                                                     0
                     0
                     1900     1950   2000     2050     2100   2150   2200         1950            2000   2050        2100        2150
                                        Wavenumber                                                       Wavenumber

                     Figure 27. Stretching modes in a-                                    Figure 28. Stretching modes in
                    Si:H. According to this double peak,                                  microcrystalline silicon – high
                       a significant part of H atoms is                                    dilution (note the low 2000 -
                    bonded as monohydide in a dense Si                                             absorption).
                                   network.

                                                                                         300


                    100                                                                  250

                                                                                         200
 Transmission (%)




                     95                                                                  150

                                                                                                         µc Si:H
                                                                                 α




                                                                                         100
                                                     Bending modes                                       (Sample T529)
                     90                                                                  50
                                              mixed a- / µc- Si:H                                        H/SiH4 = 4
                                                                                          0
                                              (sample T532)
                     85                                                                  -50
                                                                                           1900          2000               2100        2200

                            500   600   700     800      900 1000 1100                                     Wavenumber

                                        Wavenumber                                            Figure 30. Microcrystalline Silicon
                                                                                                   with low dilution ratio.
                         Figure 29. Bending mode related
                     absorption peak at around 900cm-1 in the
                             mixed state sample T532.


Figure 32 summarizes the stretching modes of samples in the amorphous and mixed state. Note
the scaling of the 2000 peak with the dilution ratio.




                                                                            31
                                                                          800
    800
                                                                                    Peaks in top-down order:
    600
                                                                                     Sample    Struct.     H/SiH4
                                                                          600        T534      a-Si        2
    400
                                                                                     T531      a-Si        2
                                                                                     T529      µc-Si       4
α




    200                                                                              T517      µc-Si       5
                                                                          400
                                                                                     T528      µc-Si       5
      0                                                                              T518      µc-Si       10
                                                                      α              T519      µc-Si       20
    -200
       1700 1750 1800 1850 1900 1950 2000 2050 2100 2150 2200 2250 2300
                                                                          200
                                  waveno



            Figure 31. Deconvolution of                                     0
              2000/2100-peak (T516).
                                                                          -200
                                                                             1400   1500   1600    1700    1800     1900   2000   2100   2200   2300

                                                                                                          wavenumber

                                                                                Figure 32. Stretching modes for several amorphous
                                                                                           and microcrystalline samples.




            6.2         Charge transport properties near and above the transition from amorphous to
                        microcrystalline silicon produced at MVSystems


            The first set of samples


            We studied the transport properties of three samples supplied by MV-Systems, Inc, Golden, CO,
            one microcrystalline silicon film (MVS723) and two amorphous silicon films, MVS722 and
            MVS823. The respective preparation parameters have not been disclosed by MVSystems at this
            time. Figures 33-35 show the respective decay curves for all three samples, whereby the a-Si:H –
            films are investigated the “classical” way as described earlier.




                                                                                              32
                                0.20                                                                              260
                                                                                                                  240
                                                                                                                  220
                                0.15                                                                              200
 Mobility (cm / V s)




                                                                                                                  180




                                                                                                  Lifetime (ns)
                                                                                                                  160
                                                                                                                  140
2




                                0.10
                                                                                                                  120
                                                                                                                  100
                                                                                                                   80
                                0.05                                                                               60
                                                                                                                   40
                                                                                                                   20
                                0.00                                                                                0
                                    0 .1                     1           10            10 0                          0.1            1          10            100
                                                          Illum ination tim e (m in)                                           Illum ination tim e (m in)


                                       1.0                                                                                  Figure 33.
        Normalized Photoconductivity




                                       0.9
                                       0.8
                                                                                                                       MVS722 (a-Si:H)
                                       0.7                                                                           Results for the mobility,
                                       0.6                                                                                lifetime, and
                                       0.5
                                       0.4
                                                                                                                    photoconductivity during
                                       0.3                                                                               light-soaking.
                                       0.2
                                       0.1
                                       0.0
                                             0.1                 1        10            100
                                                          Illum ination tim e (m in)



                                                                                                                  1000

                                 0.2
                                                                                                                   800
Mobility (cm / V s)




                                                                                               Lifetime (ns)




                                                                                                                   600
2




                                 0.1
                                                                                                                   400


                                                                                                                   200


                                 0.0                                                                                 0
                                                   0.1               1        10       100                                   0.1          1         10        100
                                                         Illumination time (min)                                                   Illumination time (min)
                                       22
         / Ω cm)




                                       20

                                       18

                                       16
                                                                                                                                        Figure 34.
      -4
          Photoconductivity (10




                                       14
                                                                                                                           Figure 4.
                                       12                                                                                     MVS823 (a-Si:H)
                                                                                                                           MVS823 (a-Si:H)
                                       10
                                                                                                                           Results for the mobility,
                                                                                                                           Decay measurement
                                        8

                                        6
                                                                                                                                 lifetime, and
                                        4                                                                                  photoconductivity during
                                        2
                                                                                                                                light-soaking.
                                        0
                                                         1               10              100

                                                          Illumination time (min)




                                                                                                                   33
                                   1.0
                                                                                                   4000
                                   0.9
                                   0.8                                                             3500
   Mobility (cm / V s)


                                   0.7                                                             3000




                                                                                   Lifetime (ns)
                                   0.6
                                                                                                   2500
   2




                                   0.5
                                                                                                   2000
                                   0.4
                                                                                                   1500
                                   0.3
                                   0.2                                                             1000

                                   0.1                                                              500
                                   0.0                                                                0
                                         1             10                 100                             1               10                 100
                                                Illumination time (min)                                            Illumination time (min)
                                   1.0
    Normalized Photoconductivity




                                   0.8

                                                                                                                     Figure 35.
                                   0.6
                                                                                                                 MVS723 (µc-Si)
                                   0.4
                                                                                                              Results for the mobility,
                                                                                                                   lifetime, and
                                   0.2                                                                          photoconductivity
                                                                                                               during light-soaking.
                                   0.0
                                         1             10                 100
                                                  Illumination time (min)




The range and depth of the long range potential fluctuations (LRPF) of these samples were
determined by the electric field dependence of the mobility and are shown in Table 4.


Table 4.                                     LRPF Range and Depth for the MVS – samples (the thickness of the intrinsic
                                             layers were about 5000Å).

                                                                Annealed State                                              Light-soaked Sate
Sample ID
                                                     Range (nm)                 Depth (eV)                         Range (nm)            Depth (eV)
MVS722 (a-Si)                                           52.9                      0.125                               51.6                  0.13
MVS823 (a-Si)                                           51.8                       0.11                              53.54                  0.13
MVS723 (µc Si)                                          400                        0.08                               400                  0.082




                                                                                                    34
The Second set of samples



The series discussed here was provided by MVSystems. It consists of eight samples starting in
the completely amorphous regime (MVS 968) in numerical order through to microcrystallinity
(MVS 975 and MVS 1003). In Table 5 the corresponding sample data is listed.



Table 5. Sample data as provided by MVSystems
             Filament                     Dark cond.       Light cond.    Thickness
Sample ID                    H2 dilution
             Temperature                  (Ωcm)-1          (Ωcm) -1
                                                                          (Å)
968          Low             0               1.0×10-10     3.1×10-5       6,000
969          High            0               1.5×10-7      1.8×10-7       70,000
970          High            20              8.5×10-7      2.5×10-6       9,200
971          High            50              9.5×10-10     1.3×10-8       19,000
972          High            80              3.0×10-8      6.8×10-8       16,000
                                                      -7            -7
973          High            90              4.3×10        4.7×10         9,000
975          High            95              2.2×10-6      2.4×10-6       7,300
1003         High            93              3.1×10-5      3.2×10-5       22,500




Figure 36 shows the decay data of the amorphous sample 968. The electric field dependent
mobility for this sample yields the range (R) and depth (D) of the Long-Range Potential
Fluctuations:

Annealed State:R = 40.4nm, D = 0.14eV
Light-soaked state: R = 38.7nm, D = 0.16eV

Unfortunately, we were unable to obtain data on most of the other samples as the photoresponse
of these is generally low and therefore not suitable for our mixing method. Only on MVS 970 we
could detect a mixing signal at an electric field of 22kVcm-1 for which we obtained a
photomixing mobility µD = 0.035cm2/Vs and a photomixing lifetime of around 60ns.




                                              35
                                               12
                Photoconductivity (10 /Ω cm)




                                               10
             -4




                                                8

                                                6

                                                4

                                                2

                                                0
                                                0.1             1            10
                                                            Illumination time (min)


                          0.10

                                                                                                            400
                          0.08
Mobility (cm / V s)




                                                                                                            300
                                                                                            Lifetime (ns)



                          0.06
2




                          0.04                                                                              200


                          0.02                                                                              100


                          0.00                                                                               0
                                                      0.1           1         10      100                         0.1           1         10      100
                                                            Illumination time (min)                                     Illumination time (min)




                                                                    Figure 36. Light-induced changes of photoconductivity,
                                                                          mobility, and lifetime for sample MVS 968




                                                                                                            36
Other three sets of new samples



In Phase II of this program, we reported our photomixing transport measurements of a series of
a-Si:H/µc-Si:H samples provided by MVSystems starting in the completely amorphous regime
through to microcrystallinity. Unfortunately, we were unable to obtain data on most of the
samples as the photoresponse of these is generally very low and therefore not suitable for our
mixing method.

We have received three series of new HWCVD H-diluted intrinsic amorphous and
microcrystalline silicon films from MVSystems for photomixing transport measurements. These
samples were deposited on glass substrates with different filament-substrate distance, substrate
temperature and H-dilution. We have performed detailed photomixing measurements on these
samples in the annealed state, and obtained interesting results, which were found to be somewhat
similar to those we reported previously on NREL HWCVD samples in the transition from
amorphous to microcrystalline silicon. While, more samples near the transition regime are
required from MVSystems in order to investigate more in detail the charge transport properties in
the transition materials.

1    Effects of the filament-substrate distance (F-S), substrate temperature and H-dilution on
    transport properties of HWCVD microcrystalline Si films.

This set of HWCVD intrinsic microcrystalline Si films were made with different filament-
substrate distance (F-S), substrate temperature and H-dilution. All samples are crystallized and of
high fraction of crystalline phase. Table 6 summarizes the growth conditions of the µc-Si:H
samples.

Table 6.   Preparation conditions and thickness of HWCVD µc-Si:H samples provided by
           MVSystems Inc.

                      Substrate       Filament-Substrate         SiH4          Thickness
 Sample
                    Temperature             Distance              (%)             (Å)
    MVS1412              High                 Long                5.06           15600
    MVS1406              High                 Long                6.25           9700
    MVS1407              High                 Long                7.41           7500


    MVS1411              Low                  Short               6.25           11200


    MVS1466              High                 Short               5.06           8400
    MVS1467              High                 Short               6.25           7500



                                                37
Let’s first see the effect of hydrogen dilution on the transport properties of µc-Si:H samples
deposited at high substrate temperature and long F-S. The results are shown in Table 7. We can
see that with increasing H-dilution up to 5.06% SiH4 in H2, the photoconductivity increases
monotonically, which is somewhat surprisingly attributed to the increase in the mobility, rather
than in the lifetime that shows little change. This interesting result is consistent with the decrease
observed in the density of charged scattering centers, which results mainly from the H-dilution-
induced decrease in the depth of the long-range potential fluctuations.


Table 7.       The effect of hydrogen dilution on the transport properties of µc-Si:H samples
               provided by MVSystems Inc.

                               Drift          Photoconducti       Depth of        Range of        VP2 /
                 Lifetime
 Sample                        Mobility       vity                LRPF            LRPF            L
                 (ns)
                               (cm2/Vs)       (S/cm)              VP (eV)         L (nm)          (×10-4)
  MVS1412
                   136.22          0.80           1.14×10-3           0.066           33.97        1.29
 SiH4: 5.06%
  MVS1406
                   173.13          0.29           8.77×10-4           0.092           39.89        2.12
 SiH4: 6.25%
  MVS1407
                   154.05          0.17           5.71×10-4           0.106           31.47        3.59
 SiH4: 7.41%



Table 8 shows the substrate temperature effect on the transport properties of µc-Si:H samples
grown at 6.25% SiH4 in H2 and short F-S. From these results, we see clearly that higher
temperature can significantly improve the film transport properties.


Table 8.       The effect of substrate temperature on the transport properties of µc-Si:H samples
               provided by MVSystems Inc.

                  Lifetime   Drift Mobility   Photoconductivity   Depth of LRPF   Range of LRPF   V P2 / L
 Sample             (ns)       (cm2/Vs)            (S/cm)            VP (eV)         L (nm)       (×10-4)
  MVS1411
                   163.00        0.23             5.41×10-4           0.099           29.26        3.34
  Tsub: Low
  MVS1467
                   546.87        0.31             3.61×10-3           0.091           50.26        1.64
  Tsub: High




Table 9 shows the influence of filament-substrate distance (F-S) on the transport properties of
µc-Si:H samples grown at high substrate temperature and two different H-dilution (5.06% and


                                                      38
6.25% SiH4 in H2, respectively). We can see that although the short F-S increases the
photoconductivity despite different degrees of H-dilution, it increases both the lifetime and
mobility for lower H-dilution due to a reduction in the charged defect density, while increases
only the lifetime for higher H-dilution.


Table 9.         The effect of flament-substrate distance on the transport properties of µc-Si:H
                 samples provided by MVSystems Inc.

                   Lifetime   Drift Mobility   Photoconductivity   Depth of LRPF   Range of LRPF   V P2 / L
 Sample              (ns)       (cm2/Vs)            (S/cm)            VP (eV)         L (nm)       (×10-4)
    MVS1412
                    136.22        0.80             1.14×10-3           0.066           33.97        1.29
    F-S: Long
    MVS1466
                    178.14        0.71             2.41×10-3           0.069           29.23        1.62
    F-S: Short


    MVS1406
                    173.13        0.29             8.77×10-4           0.092           39.89        2.12
    F-S: Long
    MVS1467
                    546.87        0.31             3.61×10-3           0.091           50.26        1.64
    F-S: Short




From the above results, we see that higher H-dilution, higher substrate temperature and short F-S
can help improve the transport properties of µc-Si:H films. While, due to fewer samples
provided, we do not observe the considerable decrease in the mobility and huge increase in the
lifetime, as we previously observed for NREL HWCVD samples in the transition from
amorphous to microcrystalline silicon when the completely microcrystalline regime is entered.



2      Effect of H-dilution on transport properties of HWCVD amorphous/microcrystalline Si
      films.


The series discussed here consists of six HWCVD intrinsic Si films. These samples were
prepared under the same conditions (middle substrate temperature and long flament-substrate
distance), except for the H-dilution, and they cover from microcrystalline Si films made with
high H-dilution to amorphous Si films made with low H-dilution. Figure 37 shows the XRD
patterns of the samples (these measurements made by Don Willimason of the Colorado School of
Mines), where the H-dilution degree and film thickness are also given. We can see that when the




                                                       39
                       H-dilution degree is over (<) ∼6% SiH4 in H2, the film becomes partially microcrystalline
                       material, which is demonstrated by considerably enhanced µc-Si:H XRD (111) and (220) peaks.

                                      (111)



                                                                                                            (220)
XRD Intensity (a.u.)




                                                                  3.85%, 8700Å



                                                                   5.66%, 7400Å

                                                                   6.25%, 8400Å


                                                                   6.83%, 8200Å

                                                                    7.41%, 9600Å




                                                                  2θ (Degree)

                               Figure 37. XRD patterns of the HWCVD intrinsic Si films samples. The H-dilution
                                                  degree and film thickness are also given.


                       Figure 38 shows the photoconductivity of the samples as a function of H-dilution. It is seen that
                       with increasing H-dilution up to 5.06% SiH4 in H2 where the microcrystalline regime has begun
                       to be entered, the photoconductivity increase dramatically, which is found to be due to a
                       considerable increase in the lifetime, as shown in Fig. 39 (a). While, the mobility does not show
                       the expected significant decrease (Fig. 39 (b)). When further increasing H-dilution to 3.85%, the
                       photoconductivity abruptly decreases by one order of magnitude mainly due to a rapid decrease
                       in the lifetime; the mobility seems to be lowered a little.

                       From the electric field dependence of the mobility, we have calculated the range and depth of the
                       potential fluctuations, and consequently estimated the relative changes in the charged defect


                                                                      40
density in the films as a function of H-dilution. The results are shown in Figures 40 and 41,
respectively. Unfortunately, we do not observe any apparent trend of changes in the depth and
range, as well as the charged defects with H-dilution ranging from 7.41% to 3.85%.

The above results indicate that as for this series of samples, perhaps neutral defects, as effective
recombination centers affecting the lifetime of carriers, play a dominant role in the charge
transport.

                                                                     -3
                                                           2.0x10
                                Photoconductivity (S/cm)


                                                                     -3
                                                           1.6x10
                                                                     -3
                                                           1.2x10
                                                                     -4
                                                           8.0x10
                                                                     -4
                                                           4.0x10
                                                                0.0
                                                                     -4
                                                           -4.0x10
                                                                          3       4                         5                    6        7        8
                                                                              SiH 4 concentration in H 2 (%)

                                                               Figure 38. The photoconductivity of the samples
                                                                         as a function of H-dilution.


                  400                                                                                                    0.32
                  350 (a)                                                                                                0.28 (b)
                                                                                               Drift mobility (cm2/Vs)




                  300                                                                                                    0.24
  Lifetime (ns)




                  250
                                                                                                                         0.20
                  200
                                                                                                                         0.16
                  150
                  100                                                                                                    0.12

                  50                                                                                                     0.08
                   0                                                                                                     0.04
                    3       4                              5    6             7       8                                      3        4       5     6      7        8
                        SiH4 concentration in H2 (%)                                                                                 SiH4 concentration in H2 (%)


                                                               Figure 39. The lifetime (a) and mobility (b)
                                                                       as a function of H-dilution.


                                                                                          41
                     0.120                                                                                    20
                              (a)                                                                             18       (b)
                     0.115




                                                                                         Range of LRPF (nm)
Depth of LRPF (eV)




                                                                                                              16
                     0.110
                                                                                                              14
                     0.105
                                                                                                              12
                     0.100
                                                                                                              10
                     0.095                                                                                        8

                     0.090                                                                                        6
                          3         4                      5        6       7       8                              3         4    5     6      7       8

                               SiH4 concentration in H2 (%)                                                             SiH4 concentration in H2 (%)

                                                         Figure 40. The depth (a) and range (b) of potential
                                                              fluctuations as a function of H-dilution.


                                                         1.2

                                                         1.1
                                    N ∝ VP / L (x 10 )
                                    -3




                                                         1.0

                                                         0.9

                                                         0.8
                                    2




                                                         0.7

                                                         0.6

                                                         0.5
                                                               3        4       5                             6              7    8
                                                                   SiH4 concentration in H2 (%)

                                                               Figure 41. The relative change in the density of
                                                                 charged defects as a function of H-dilution.




                                                                                    42
3. Effect of H-dilution on transport properties of HWCVD amorphous/microcrystalline Si
   films prepared under optimized conditions.


We have studied above the transport properties of HWCVD amorphous/microcrystalline Si films
with respect to deposition parameters, such as filament-substrate distance (F-S), substrate
temperature and H-dilution. We have found that higher H-dilution, higher substrate temperature
and short F-S can improve the transport properties of films. While, we do not see yet well-
behaved properties near the amorphous to microcrystalline transition regime. Fortunately, we
have received the third set of HWCVD samples from MVSystems Inc. These samples were
grown under optimized conditions (short filament-substrate distance and high substrate
temperature) with a little wider range of H-dilution. This set of samples may allow us to
investigate more in detail the charge transport properties in the transition materials. Table 10
summarizes the H-dilution used (silane concentration in H2) and the film thickness.


Table 10. Preparation conditions and thickness of HWCVD µc-Si:H samples provided by
          MVSystems Inc.

        Sample               SiH4 (%)       Thickness (Å)
           MVS1480              8.75            8200
           MVS1481              8.75            11000
           MVS1482              5.00            8200          Short filament-
           MVS1483              7.50            11300         substrate (F-S)
                                                                 distance
           MVS1484              6.25            9800
           MVS1485             10.00            11300
                                                               High substrate
           MVS1486              3.75            9300
                                                                temperature
           MVS1487              7.00            8800
           MVS1488              8.00            9900
           MVS1489              9.50            7800
           MVS1490              7.50            13000



Figure 42 shows the photoconductivity of the samples as a function of H-dilution. It is seen that
the photoconductivity does not change monotonically with H-dilution. With increasing H-
dilution up to about 7-8% SiH4 in H2, the photoconductivity surprisingly decrease at first, but
with further increasing H-dilution, it begins to increase considerably; when H-dilution is
increased over 4%, the photoconductivity is increased by about two orders of magnitude. This
non-monotonic change is attributed to the similar changes in both the mobility and lifetime with
H-dilution (see Fig. 43), in contrast to the case of the samples prepared at middle substrate
temperature and long F-S distance.


                                               43
The non-monotonic change observed particularly in the mobility has been confirmed by the
determination of the depth and range of the potential fluctuations, and subsequently the relative
changes in the charged defect density in the films as a function of H-dilution. From Fig. 44, we
see that the depth peaks at medium H-dilution (7-8%), while decreases rapidly when H-dilution
is over 4%, which leads to similar changes in the charged defect density shown in Fig. 45.

The above results suggest that both charged defects and neutral defects may be involved in the
charge transport of the third set of films.

From the photomixing transport results of these three series of HWCVD
amorphous/microcrystalline Si samples, we have seen some different behaviors with H-dilution.
This might be due to different microstructure in the films. Therefore, to further understand the
transport properties, it is necessary to perform detailed studies of structural properties of the
samples prepared at different conditions, particularly across the entire range of H-dilution (0-
100%). We anticipate more samples near the transition regime are available from MVSystems in
order to investigate more in detail the charge transport properties in the transition materials.
           Photoconductivity (S/cm)




                                           -3
                                      10


                                           -4
                                      10


                                           -5
                                      10


                                           -6
                                      10
                                                3    4    5   6   7    8   9   10    11
                                                    SiH 4 concentration in H 2 (%)

          Figure 42. The photoconductivity of the samples as a function of H-dilution.




                                                                  44
                                      0.16                                                                                          24

                                      0.14                                                                                          20




                                                                                                               Range of LRPF (nm)
                 Depth of LRPF (eV)

                                      0.12                                                                                          16

                                      0.10                                                                                          12

                                      0.08                                                                                          8

                                      0.06                                                                                          4
                                                                                                   (a)                                                                                       (b)
                                      0.04                                                                                          0
                                          3       4       5       6       7       8       9    10        11                          3       4       5       6       7       8       9       10    11

                                                  SiH4 concentration in H2 (%)                                                               SiH4 concentration in H2 (%)




                                                  Figure 43. The mobility(a) and lifetime (b) as a function of H-dilution.




                                      1
                                                                                              (a)                                                                                                 (b)
Drift mobility (cm2/Vs)




                                                                                                                                    100
                                                                                                               Life time (ns)




                              0.1


                                                                                                                                     10


                                      3       4       5       6       7       8       9       10     11                                  3       4       5       6       7       8       9        10    11

                                              SiH4 concentration in H2 (%)                                                                    SiH4 concentration in H2 (%)

                                          Figure 44. The depth (a) and range (b) of potential fluctuations as a function of H-
                                                                              dilution.




                                                                                                              45
                     12


N ∝ VP / L (x 10 )
-4                   11

                     10

                     9
2




                     8

                     7

                     6
                          3    4     5      6        7   8     9     10      11
                              SiH4 concentration in H2 (%)


                          Figure 45. The relative change in the density of
                             charge defects as a function of H-dilution.




                                                46
6.3   Charge transport properties of microcrystalline silicon prepared by Pulsed PECVD
      technique (MVSystems Inc.)


During the third quarter of Phase III, We have conducted detailed photomixing transport
measurements on three series of new HWCVD H-diluted intrinsic amorphous and
microcrystalline silicon films supplied by MVSystems. These samples were deposited on glass
substrates with different filament-substrate distance, substrate temperature and H-dilution. It was
found that higher H-dilution, higher substrate temperature and short F-S help improve the film
transport properties. The results were found to be somewhat similar to those we reported
previously on NREL HWCVD samples in the transition from amorphous to microcrystalline
silicon. While, more samples near the transition regime are required from MVSystems in order to
perform well-defined studies of the charge transport properties in the transition materials.

We have received a series of microcrystalline silicon films prepared by the Pulsed PECVD
technique from MVSystems for photomixing transport measurements. These samples were
deposited on glass substrates with varying H-dilution and substrate temperature. They made n/i/p
solar cells using the same recipe for some of the respective films. This may allow us to correlate
intrinsic film properties with the related device performance.

Table 11 shows the deposition conditions and some measured properties of the microcrystalline
Si samples, as disclosed by MVSystems. The samples nrct#149-154 are a series of µc-Si:H films
deposited with varying H-dilution, and the samples nrct#175, 176, and 181 are µc-Si:H films
deposited with differing substrate temperature.

Table 12 shows the XRD results (Don Williamson, Colorado School of Mines) of the µc-Si:H
samples. The quantity R(220) is the ratio of the (220) XRD peak area to the (111) peak area. In
powder Si, this ratio should be 55%; for the films near the amorphous-to-microcrystalline
transition, it is seen to be in the range of 700%, indicating that the transition films are strongly
(220) oriented. We can see from Table 12 that H-dilution has different effects on the (111) and
(220) crystallites. Increasing H-dilution reduces the (111) grain size, but increases the (220)
grain size. Furthermore, the ratio of the (220) peak area to the (111) peak area does not change
monotonically with increasing H-dilution; it increases up to a maximum (740%) at H-dilution
degree of 93.6%, but with further increasing H-dilution, it begins to decrease again. These results
indicate that at lower H-dilution, particularly 93.6%, the films are strongly (220) oriented, and
near the transition regime. When H-dilution increases beyond about 96%, the films change to
dominantly (111) oriented although the (220) grain size increases.

From the XRD results of samples deposited at H-dilution degree of 93.6% with differing
substrate temperature, we see that higher substrate temperature favors the microcrystalline
formation. Increasing substrate temperature increases both the (111) and (220) grain size, as well
as the R(220) ratio. When the substrate temperature is higher than 400ºC, the films are strongly
(220) oriented, indicating near the transition regime; but for film nrct#176 deposited at 330ºC,
there is no microcrystallites detectable by XRD, and also it has almost no photoresponse,
suggesting very poor transport properties.




                                                47
Table 11. Deposition conditions and properties of Pulsed PECVD µc-Si:H samples provided
          by MVSystems.

                                                                                                Activation
              Temperature    Pressure   H Dilution    Thickness    Light Cond.     Dark Cond.
    SN                                                                                           Energy
                 (ºC)        (mTorr)       (%)           (Å)          (S/cm)         (S/cm)
                                                                                                  (eV)
 nrct#149         410          400         97.3           6000         5.40×10-5    9.10×10-6      —
 nrct#150         410          400         96.1           4900         2.30×10-4    2.20×10-5      —
 nrct#152         410          400         94.8           6000         1.60×10-4    1.10×10-5      —
 nrct#153         410          400         93.6           7300         8.00×10-5    5.50×10-6      —
 nrct#154         410          400         92.4           7300         6.00×10-5    6.00×10-6      —

 nrct#175         410          200         93.6           6700         2.90×10-5    2.80×10-6      0.48

 nrct#176         330          200         93.6           4850         3.40×10-7    3.20×10-7      0.27

 nrct#181         490          200         93.6           3600         2.90×10-5    7.00×10-7      0.53




Table 12. XRD characterizations of Pulsed PECVD µc-Si:H samples provided by
          MVSystems.

         SN             (111) grain size (Å)         (220) grain size (Å)           R (220) (%)
    nrct#149                    180                              280                     62
    nrct#150                    270                              250                     124
    nrct#152                    270                              350                     185
    nrct#153                    250                              220                     740
    nrct#154                    270                              220                     370

    nrct#175                    210                              340                     600
    nrct#176                None visible                  None visible                   —
    nrct#181                    230                              350                     650


Figures 46-49 show the effects of H-dilution on photomixing transport properties of Pulsed
PECVD µc-Si:H samples. From Fig.46, we see that with increasing H-dilution near the transition
regime (about 93.6%), the photoconductivity decreases, which is found to be due to a decrease in
the drift mobility (Fig.47(a)), while there is little change in the lifetime (Fig.47(b)). With



                                                     48
increasing H-dilution beyond the transition regime, the photoconductivity begins to increase
quickly, but then remains almost constant with further increasing H-dilution. This increase
observed in the photoconductivity beyond the transition regime results from a considerable
increase in the lifetime although the mobility is reduced, as verified by a decrease in the depth of
the potential fluctuations with increasing H-dilution (Fig.48(a)). Furthermore, it is interesting to
observe when H-dilution is increased beyond the transition regime so that the microcrystalline
regime is entered, the range of the potential fluctuations is increased considerably (Fig.48(b)),
indicating more ordering in the Si network structure. A longer range is correlated with a lower
concentration of charged defects as scattering centers. As shown in Fig.49, The charged defect
density decreases a little when H-dilution is increased near the transition regime, and beyond the
transition regime, it decreases considerably.




                                                10
                                                 9
                 Photoconductivity (x10 S/cm)




                                                 8
                                                 7
                                                 6
                 -4




                                                 5
                                                4

                                                3


                                                2




                                                1
                                                 92   93   94    95    96   97   98

                                                           H dilution (%)

                Figure 46. The effect of H-dilution on the photoconductivity of the
                                             samples.




                                                                 49
                          0.25                                                                           800
                                   (a)                                                                              (b)
                                                                                                         700
                          0.20
Drift mobility (cm2/Vs)




                                                                                                         600

                                                                                                         500




                                                                                         Lifetime (ns)
                          0.15
                                                                                                         400
                          0.10                                                                           300

                                                                                                         200
                          0.05
                                                                                                         100

                          0.00                                                                                0
                              92         93   94    95   96    97    98                                        92         93        94        95        96        97        98

                                              H dilution (%)                                                                        H dilution (%)



                                     Figure 47. The effect of H-dilution on the mobility (a) and lifetime (b).




                                                                                                         3
                          0.18                                                                      10
                                   (a)                                                                        (b)
                          0.16
                                                                               Range of LRPF (nm)
Depth of LRPF (eV)




                          0.14
                                                                                                         2
                                                                                                    10
                          0.12


                          0.10


                                                                                                         1
                          0.08                                                                      10
                              92         93   94    95    96   97    98                                  92         93         94        95        96        97        98

                                              H dilution (%)                                                                   H dilution (%)


                                                     Figure 48. The effect of H-dilution on the depth (a)
                                                           and range (b) of potential fluctuations.



                                                                          50
                                    5


                                    4
               N ∝ VP / L (x 10 )
               -4




                                    3


                                    2
               2




                                    1


                                    0
                                     92      93      94       95      96      97      98
                                                     H dilution (%)


                                    Figure 49. The effect of H-dilution on the relative change
                                                in the density of charged defects.

Figures 50-53 show the effects of substrate temperature on photomixing transport properties of
Pulsed PECVD µc-Si:H samples deposited near the transition regime ( 93.6% H-dilution). It is
seen that when the substrate temperature is lower than 400ºC, the transport properties are very
poor, as the photoconductivity is more than one order of magnitude smaller (Fig.50) due to very
low mobility (Fig.51(a)). This results in failure to determine the range and depth of the potential
fluctuations. With increasing the substrate temperature up to 410ºC, the lifetime is increased a
little, but the mobility is increased significantly, by about one order of magnitude, leading to
much enhanced photoresponse. Further increasing the substrate temperature continues to
increase the lifetime considerably, but reduces the mobility too, indicating a non-monotonic
change in the mobility with substrate temperature. This results in almost constant
photoconductivity at higher substrate temperature than 400ºC. It is note-worthy that although
higher substrate temperature reduces the mobility due to an increase in the depth of the potential
fluctuations (Fig.52(a)), it dramatically increases the range (Fig.52(b)), leading to much lower
density of charged defects in the films (Fig.53).




                                                              51
n/i/p solar cells have been made by MVSystems using the same recipe for films #175 and #181.
The latter film deposited at higher substrate temperature appears to result in an improved solar
cell, compared to the former. This indicates that there is a scaling between the intrinsic film
properties and the related device performance, and higher substrate temperature is required for
the Pulsed PECVD technique to produce high quality films and related devices near the
transition regime.                   Photoconductivity (x10 S/cm)



                                                                     10
                                     -4




                                                                      1




                                                                    0.1
                                                                      300          350             400                     450             500
                                                                                                                                 o
                                                                             Substrate temperature ( C)
                                                                            Figure 50. The effect of substrate temperature on the
                                                                                     photoconductivity of the samples.


                                  Figure 51. The effect of substrate temperature on the mobility (a) and lifetime (b).

                                                                                                                           500
                                   (a)                                                                                               (b)
                                                                                                                           400
 Drift mobility (cm /Vs)
 2




                            0.1
                                                                                                           Lifetime (ns)




                                                                                                                           300


                                                                                                                           200


                                                                                                                           100
                           0.01

                                                                                                                             0
                             300                                    350      400         450         500                     300           350   400      450       500
                                                                                               o                                                                o
                                     Substrate temperature ( C)                                                                       Substrate temperature ( C)


                                                                                                           52
                                                                                                3
                     0.18                                                                     10
                            (a)                                                                       (b)
                     0.16




                                                                         Range of LRPF (nm)
Depth of LRPF (eV)   0.14
                                                                                                2
                                                                                              10
                     0.12


                     0.10


                                                                                                1
                     0.08                                                                     10
                        300       350      400      450            500                          300         350    400      450       500
                                                                                                                                  o
                               Substrate temperature ( C)
                                                          o
                                                                                                        Substrate temperature ( C)

                            Figure 52. The effect of substrate temperature on the depth (a)
                                        and range (b) of potential fluctuations.




                                                              53
                                         5


                                         4


                    N ∝ VP / L (x 10 )
                   -4                    3


                                         2
                   2




                                         1


                                         0
                                         300     350       400         450        500
                                                                              o
                                               Substrate temperature ( C)
                                    Figure 53. The effect of substrate temperature on
                                   the relative change in the density of charged defects.

7. High Deposition Rate Preparation of a-Si:H by HWCVD

7.1   First series of samples

The deployment of solar cells for large scale power generation is contingent not only upon the
production of photovoltaic devices of stable high efficiency but for industrial production, high
deposition rate is desirable to improve the throughput for a given machine, and also to reduce the
costs and the capital investment. It has been suggested that with increasing deposition rates,
higher silane related radicals, short-lifetime radicals, increased ion-bombardment energies, and
hence more resultant microvoids and defects, are potential causes for deterioration of the
performance and stability of hydrogenated amorphous silicon (a-Si:H) films. Various techniques
to improve the deposition rate have been attempted: with plasma enhanced chemical vapor
deposition (PECVD) which usually yields material with poor initial performance and poor
stability when deposition rates are increased, which has been correlates with the increased
density of microvoids [19]. A correlation between the gas phase species in the silane plasma and
the properties of a-Si:H films deposited by PECVD has indicated that the higher-order silane
species contribute to the cause of light-induced degradation in the film quality at high deposition
rates [20]. The growth rate of these a-Si:H films ranged from 2 Å/s to 20 Å/s.

At present, it is generally believed that the performance of a-Si:H will deteriorate monotonically
with increasing deposition rates, but the high deposition rates achieved by PECVD are relatively
low, less than 20 Å/s. We have performed photoconductive frequency mixing measurements on
much higher deposition rate HWCVD a-Si:H samples provided by Brent Nelson of NREL, and


                                                          54
found some interesting results. The charge transport properties of the samples do not change
monotonically with increasing deposition rates.

Brent Nelson has employed the HWCVD technique to prepare a-Si:H films grown with
deposition rates ranging from 32 to 191 Å/s. He achieved the high deposition rates by making the
following changes to one of his HWCVD reactors.


Deposition Parameter                   Standard Config.          High Dep. Rate Config.

Number of filaments                            1                              2
Filament to substrate spacing                5 cm                          3.2 cm
Silane flow rate                          20-50 sccm                    50-100 sccm
Deposition Pressure                      10-20 mTorr                   20-120 m Torr

Brent Nelson has informed us: “The deposition rate is a complicated function of the pressure,
silane flow, and filament current. These are the three main parameters we change to alter the
deposition rate. The increase in silane flow is necessary to accommodate the increased silane
depletion due to multiple filaments. The increased pressure is necessary.” We were supplied with
the following samples whose preparation parameters are indicated in the Table 13.

Table 13. Growth conditions and properties of high deposition rate HWCVD samples.

 Sample Temp.  Press. F(SiH4)             I(fil.)     Thk.    Rate      Cond.        H(IR)
   ID    (ºC) (mTorr) (sccm)               (A)        (Å)     (Å/s)     Rate         (at.%)
  L079       295       20         20        30         6644    31.64   1.29×104       4.51
  L077       295       70         20        30         8530    47.39   1.17×104       4.81
                                                                                 4
  L110       285       19         75        30        10514    58.41   8.60×10        9.50
                                                                                 5
  L139       315       50         50        29         8144    67.87   1.15×10        7.10
  L169       318       50         50        30         9235    76.96   1.02×105       6.50
                                                                                 5
  L104       350       50         50        30        10179    84.83   1.11×10        7.50
                                                                                 4
  L183       348       70         75        30        11000   109.78   5.72×10        5.60
  L178       317       70         75        30        11381   113.58   2.63×104       6.50
                                                                                 4
  L182       348       70         75        30        11502   114.79   8.63×10        4.90
                                                                                 4
  L190       326       50         75        30         9318   132.74   5.64×10        6.70
  L196       325       75         75        30         8538   142.30   2.30×104       5.80
                                                                                 4
  L094       385       90         75        30        13701   152.23   2.37×10        6.79
                                                                                 2
  L194       325       70        100        30        10662   177.70   2.64×10        5.00
  L087       295       120        75        30        22900   190.83     5.73          —




                                                 55
We have characterized these samples in both the annealed and light-soaked states utilizing the
photoconductive frequency mixing technique enabling us to determine the drift mobility, lifetime
and from the electric field dependence of these quantities, we have been able to determine the
range and the depth of the long range potential fluctuations and from the latter two quantities we
were able to determine the relative differences in the charge centers which exist in the samples.

Measurements of transport properties of high deposition rate HWCVD a-Si:H films in the
  annealed state

The photoconductivity as a function of deposition rate is shown in Figure 54. It is seen that in the
deposition range from 30 Å/s to ~150 Å/s, the photoconductivity hovers at levels of 10-4 (S/cm)
except that in the neighborhood of the deposition rate of 70 Å/s, the photoconductivity peaks in
the neighborhood of 10-2 (S/cm), two orders of magnitude increase above the average! Above the
deposition rate of ~150 Å/s, the photoconductivity starts to plummet to 10-8 (S/cm)!

By measuring the photomixing signal at 250 Mhz in conjunction with the dc photoconductivity,
we determined the drift mobility and the lifetime. Figure 55 shows the drift mobility as a function
of deposition rate. Where it is seen that the drift mobility peaks in the neighborhood of 70 Å/s
where the photoconductivity peaks. However, it should be noted that the drift mobility varies by
integer values, while the photoconductivity varies by orders of magnitude. These differences can
be accounted for by the fact that the lifetime as a function of deposition rate varies by orders of
magnitude as can be seen in Figure 56.

                                                     -1
                                                10
                                                     -2
                     Photoconductivity (S/cm)




                                                10
                                                     -3
                                                10
                                                     -4
                                                10
                                                     -5
                                                10
                                                     -6
                                                10
                                                     -7
                                                10
                                                     -8
                                                10
                                                          30   60   90        120   150   180   210
                                                               Deposition rate (Å/s)

                 Figure 54. The photoconductivity as a function of deposition rate.




                                                                         56
                                                      0.5
                                                                                                                                                  3
                                                                                                                                  10
                            Drift mobility (cm /Vs)



                                                      0.4
                            2




                                                                                                            Life time (ns)
                                                      0.3                                                                                         2
                                                                                                                                  10
                                                      0.2

                                                                                                                                                  1
                                                      0.1                                                                         10

                                                      0.0
                                                                                                                                                            30        60     90   120   150   180   210
                                                                30    60    90    120   150    180    210
                                                                                                                                                                      Deposition rate (Å/s)
                                                                       Deposition rate (Å/s)


                                                        Figure 55. The drift mobility as a                                Figure 56. The lifetime as a function
                                                           function of deposition rate.                                            of deposition rate.
Figure 57 shows a typical electric field dependence of the drift mobility (a) and lifetime (b).
From this type of data, we have calculated the range and depth of the long range potential
fluctuations which are shown in Figure 58. It should be observed that minimal depth of potential
occurs in the region where the drift mobility and photoconductivity peaks as is to be expected.

                                    0.32                                                                                                          0.20
                                                                                                                                                                                                    (a)
                                                                                                                                                  0.18
                                    0.28
  Drift mobility (cm /Vs)




                                                                                                                        Depth of LRPF (eV)




                                                                                                                                                  0.16
                                    0.24
 2




                                                                                                                                                  0.14
                                    0.20
                                                                                                                                                  0.12
                                    0.16                                                                                                          0.10

                                    0.12                                                                                                          0.08
                                                                                                (a)
                                    0.08                                                                                                          0.06
                                                            0    2000 4000 6000 8000 10000 12000                                                                 30        60     90    120   150    180

                                                                     Electric field (V/cm)                                                                                 Deposition rate (Å/s)
                                        180                                                                                                           120
                                                                                               (b)                                                                                                  (b)
                                        160                                                                                                           100
                                                                                                                             Range of LRPF (nm)




                                        140                                                                                                           80
          Life time (ns)




                                        120                                                                                                           60
                                        100
                                                                                                                                                      40
                                                80
                                                                                                                                                      20
                                                60
                                                                                                                                                       0
                                                40
                                                            0    2000 4000 6000 8000 10000 12000                                                                 30        60     90    120   150    180
                                                                     Electric field (V/cm)                                                                                 Deposition rate (Å/s)
 Figure 57. Field dependence of the drift                                                                                        Figure 58. The depth (a) and range (b) of
mobility (a) and lifetime (b) for sample L183. 57                                                                                the long range potential fluctuations as a
                                                                                                                                        function of deposition rate.
It can be shown that the density of charge defects (n) is given by:


                                                            n ∝ VP 2 /L                                                                     (1)

Where VP and L represent the depth and range of the long range potential fluctuations,
respectively.

Figure 59 shows the relative change in the charged defects as a function of deposition rate. It can
be seen that the charged defect density remains constant up to the deposition rate of ~150 Å/s, but
above, it increases considerably.



                 1.8                                                                          12
                 1.6
                                                                                              10
                 1.4

                                                                    Hydrogen content (at.%)
VP / L (x 10 )
-3




                 1.2                                                                           8
                 1.0
                 0.8                                                                           6
                 0.6
2




                                                                                               4
                 0.4
                 0.2                                                                           2
                 0.0
                 -0.2                                                                          0
                        30   60   90   120   150     180   210                                     30   60   90   120   150     180   210
                             Deposition rate (Å/s)                                                      Deposition rate (Å/s)


  Figure 59. The relative change in the charged                                                Figure 60. The hydrogen content as
     defects as a function of deposition rate.                                                    a function of deposition rate.

It is of great interest to ascertain what are the chemical or structural properties of the films that
vary as a function the deposition rate. The hydrogen content of the samples as a function of
deposition rate is shown in Figure 60. It is interesting to note that maximal hydrogen content
appears to occur in the region where the drift mobility and photoconductivity peaks, while
elsewhere, the hydrogen content is comparable, about 6 at.%. It is clear that it is necessary to
look for other film properties that are responsible for variation of charge transport properties as a
function of deposition rate. It would be of interest to determine the microvoid and higher order
silane content by performing SAXS and IR measurements in the Si-H stretching mode region.

Brent Nelson has provided us some preliminary IR results. There is no increase in dihydride
bonding with increasing deposition rate. Table II shows qualitative results of SiH2 shoulder at
2009 cm-1 from the IR spectra. We found that both dark-conductivity and photoconductivity of
the samples with detectable SiH2 shoulder are smaller than those of other samples without
detectable SiH2 shoulder, indicating poorer performance. In addition, we anticipate David
Cohen’s DLCS measurements of defect density are available.



                                                                  58
Light-induced decay measurements of transport properties of high deposition rate HWCVD
a-Si:H films

Figure 61 shows the normalized photoconductivity (a), drift mobility (b) and lifetime (c),
respectively, for three typical high deposition rate HWCVD a-Si:H samples as a function of
illumination time.
                Photoconductivity (S/cm)


                                                                                      (a)
                                             1
                                           0.9
                                           0.8
                                           0.7
                                           0.6
                                           0.5   L079 (31.64 Å /s)
                                           0.4   L182 (114.79 Å /s)
                                                 L094 (152.23 Å /s)
                                           0.3
                                                      0                 1         2
                                                 10                10        10
               Drift mobility (cm /Vs)
              2




                                             1
                                           0.9
                                           0.8
                                           0.7
                                           0.6   L079 (31.64 Å /s)
                                           0.5   L182 (114.79 Å /s)
                                                 L094 (152.23 Å /s)
                                                                                      (b)
                                           0.4
                                                      0                 1         2
                                                 10                10        10

                                                                                      (c)
                                             1
                                           0.9
               Life time (ns)




                                           0.8
                                           0.7
                                           0.6
                                           0.5
                                                 L079 (31.64 Å /s)
                                           0.4   L182 (114.79 Å /s)
                                                 L094 (152.23 Å /s)
                                           0.3
                                                      0                 1         2
                                                 10                10        10
                                                      illum ination tim e (m in.)
       Figure 61. Normalized photoconductivity, drift mobility and lifetime for high
          deposition rate HWCVD a-Si:H samples as a function of illumination.

                                                              59
It is seen that with light-soaking, the photoconductivity for all the samples decreases, which is
demonstrated to be attributed to considerable decreases in the lifetime. It is interesting to note
that the drift mobility changes very small, even a little increase upon light-soaking!

In order to investigate more in detail the influence of illumination on the transport properties of
the samples as a function of deposition rate, we performed systematical measurements on all
samples whose deposition rates range from 32 to 191 Å/s. The photoconductivity, drift mobility
and lifetime as a function of deposition rate in both the annealed and the light-induced states are
shown in Figure 62 (a), (b) and (c), respectively. It can be seen that light-soaking reduces the
photoconductivity and the lifetime, as observed before, while surprisingly increases the mobility
for all the samples. Furthermore, the deposition rate dependence of these transport parameters in
the light-soaked state is similar to that in the annealed state. Since the mobility varies only by
integer values, while the lifetime varies by orders of magnitude, the light-induced decrease in the
photoconductivity is mainly due to the decrease in the lifetime.



                                                                                                             0.6

                                                                                                             0.5   (b)                               State A
                                                                                                                                                     State B
                                                                                   Drift mobility (cm2/Vs)

                                                                                                             0.4

                                                                                                             0.3
                              -2
                             10                                                                              0.2
                                                                       (a)
  Photoconductivity (S/cm)




                              -3
                             10                                                                              0.1

                              -4
                                                                                                             0.0
                             10
                                                                                                                   30      60      90   120    150   180       210
                              -5
                             10                                                                                              Deposition rate (Å/s)

                              -6
                             10
                                           State A                                                           10
                                                                                                               3                                        (c)
                              -7
                             10            State B
                                                                                    Life time (ns)




                              -8
                             10
                                     30    60        90   120   150   180    210                               2
                                                                                                             10
                                            Deposition rate (Å/s)

                                                                                                               1         State A
                                                                                                             10
                                                                                                                         State B

                                                                                                                             70               140              210
                                                                                                                           Deposition rate (Å/s)

                                   Figure 62. The light-induced changes in the photoconductivity, drift mobility and
                                                        lifetime as a function of deposition rate.


                                                                             60
From the electric field dependence of the drift mobility and lifetime, we have calculated the
depth and range of the long-range potential fluctuations. Figure 63 shows the effect of light-
soaking on the depth (a) and range (b) of potential fluctuations as a function of deposition rate.
We see that illumination reduce considerably the range, as should be expected, but meanwhile
also reduce the depth a little which verify the light-induced increase in the mobility observed
above.



                     0.20                                                                       120
                                         State A               (a)                                         State A                     (b)
                     0.18                                                                       100
                                         State B                                                           State B




                                                                           Range of LRPF (nm)
Depth of LRPF (eV)




                     0.16                                                                       80
                     0.14                                                                       60
                     0.12
                                                                                                40
                     0.10
                                                                                                20
                     0.08
                                                                                                 0
                     0.06
                            30      60       90    120   150    180                                   30       60    90    120   150   180
                                   Deposition rate (Å/s)                                                      Deposition rate (Å/s)

                                 Figure 63. The light-induced changes in the depth and range of potential
                                              fluctuations as a function of deposition rate.




Figure 64 shows the light-induced relative change in the density of charged defects as a function
of deposition rate. We see that the charged defect density increases upon light-soaking for all
deposition rates. It is interesting to note that there appears to be the minimum light-induced
increase at the intermediate deposition rates where the photoconductivity, drift mobility and
lifetime peak. In addition, it is generally expected that due to the increase in the density of
charged defects during the light-soaking process, the depth of the potential fluctuations has a
tendency to increase, whereas the range of the potential fluctuations has a tendency to decrease.
While, this is not the case here. The light-induced increase in the density of charged defects
results from the much more decrease in the range than in the depth.




                                                                      61
                     5                 10
                                        9
                                        8                                                   State A
                                        7
                     4   nLS / nAN (a.u.)                                                   State B
n ∝ VP / L (x 10 )

                                        6
-3




                                        5
                                        4
                                        3
                                        2
                     3                  1
                                        0
                                             30   60    90   120   150    180   210

                                                  Deposition rate (Å/s)
2




                     2

                     1

                     0

                                            30         60          90           120   150    180      210
                                                       Deposition rate (Å/s)

            Figure 64. The light-induced relative change in the charged defect density
                                  as a function of deposition rate.




                                                                           62
7.2   The second series of samples

Electronic Transport Properties of High Deposition Rate HWCVD a-Si:H in the annealed
state.

Recently, Brent Nelson et al. at NREL have employed the HWCVD technique to prepare a-Si:H
films grown with deposition rates up to 1 µm/min [21]. They found that the deposition rate
increases with increasing deposition pressure, silane flow rate, and filament current and
decreasing filament-to-substrate distance. There are significant interactions among these
parameters that require optimization to grow films of optimal quality for a desired deposition
rate. An AM1.5 photoconductivity-to-dark-conductivity ratio of 105 could be maintained at
deposition rates up to 130 Å/s under the best conditions, beyond which the conductivity ratio
decreases. Other electronic properties, e.g. the ambipolar diffusion length, Urbach energy, and
as-grown defect density, decrease more rapidly with increasing deposition rate,. The density of
microvoids determined by small-angle X-ray scattering (SAXS) was found to increase by well
over an order of magnitude when going from one to two filaments. However, both Raman and X-
ray diffraction (XRD) measurements show no change in film structure with increasing deposition
rates up to 144 Å/s, and atomic force microscopy (AFM) reveals little change in topology.

In phase II of this program, we have reported the photomixing measurements of the first set of
high deposition rate HWCVD a-Si:H films supplied by Brent Nelson. The deposition rates range
from 32 to 191 Å/s. Shortly thereafter, we received the second set of high deposition rate a-Si:H
samples. These new samples were deposited on two kinds of substrates; a 1737 glass substrate
for us and a stainless steel substrate for David Cohen at the University of Oregon to determine
the defect density by the drive level capacitance spectroscopy (DLCS). This new series has a
couple of advantages over the first series of samples. First, it allows us to compare different
characterization techniques used on samples grown simultaneously as a part of the same series.
Second, some of the growth conditions have been optimized a little further than the first series.
He achieved the high deposition rates by making the following changes to one of his HWCVD
tube-reactors.


Deposition Parameter              Standard Configuration         High Dep. Rate Config.
Number of filaments                          1                              2
Filament to substrate spacing              5 cm                          3.2 cm
Silane flow rate                        20-50 sccm                     40-75 sccm
Deposition pressure                    10-20 mTorr                    30-70 m Torr
Typical deposition rates                 5-20 Å/s                      40-150 Å/s


Brent Nelson has informed us: “The increase in silane flow is necessary to accommodate the
increased silane depletion due to multiple filaments. The pressure is increased to achieve an
optimal number of gas phase collisions at closer filament spacing, and the filament currents used
are similar to those for the normal one filament HWCVD process in order to maintain the correct
reactions with the silane and the W filament” We were supplied with the following samples
whose preparation parameters are indicated in the Table.14.



                                               63
Table 14. Growth conditions and properties of high deposition rate HWCVD samples.

  Sample     Temp.      Press.     F(SiH4)     Thk.       Rate     Photo Cond.       E04
    ID        (ºC)     (mTorr)     (sccm)       (Å)       (Å/s)        (S/cm)        (eV)
   L256       316         30          40       19715       50        2.36×10-4       1.83
   L253       316         35          50       19266       70        2.64×10-4       1.83
   L252       316         50          50       19084       90        3.71×10-4       1.84
   L254       325         50          75       19346       110       2.57×10-4       1.84
   L255       325         70          75       20375       130       1.61×10-4       1.82

We have characterized these new samples in the annealed state utilizing the photoconductive
frequency mixing technique enabling us to determine the drift mobility, lifetime and from the
electric field dependence of these quantities, we have been able to determine the range and the
depth of the long range potential fluctuations and from the latter two quantities we were able to
determine the relative differences in the charge centers which exist in the samples.

The photoconductivity as a function of deposition rate is shown in Figure 65. It is seen that in the
deposition range from 50 to 130 Å/s, the photoconductivity hovers at levels of 10-4 (S/cm), similar
to the case of the first set of samples. While, in the neighborhood of the deposition rate of 90 Å/s,
the photoconductivity also peaks around 10-4 (S/cm), not as large as for the first series samples.

We determined the drift mobility and the lifetime by measuring the photomixing signal at 250
MHz in conjunction with the dc photoconductivity. Figures 66 and 67 show the lifetime and the
drift mobility as a function of deposition rate, respectively. It is seen from the figure 2 that the
lifetime peaks in the neighborhood of 90 Å/s where the photoconductivity peaks. However, it
should be noted that there is a minimal value of the drift mobility in the neighborhood of 90 Å/s,
just opposite to the case of the first series samples!

Figure 68 shows the electric field dependence of the drift mobility (a) and the lifetime (b) as a
function of deposition rate. From this type of data, we have calculated the range and depth of the
long range potential fluctuations which are shown in Figure 69. Surprisingly, it is observed that
both the depth and the range of potential fluctuations peak in the region where the drift mobility
shows a minimal value. This leads us to believe that the changes in photoconductivity as a
function of deposition rate are dominantly attributed to the changes in the lifetime, rather than in
the drift mobility as observed for the first series samples. Actually, from the figures 66 and 67,
we see clearly that the lifetime varies by about one order of magnitude, while the drift mobility
varies only by integer values.




                                                64
                                                                     4.0




                                       Photoconductivity (10 S/cm)
                                                                     3.5

                                       -4                            3.0

                                                                     2.5

                                                                     2.0

                                                                     1.5

                                                                     1.0
                                                                       40    60        80                     100               120         140
                                                                                  Deposition rate (Å/s)

                            Figure 65. The photoconductivity as a function of deposition
                                                     rate.


                                                                                                                       0.25

                  3
                 10                                                                                                    0.20
                                                                                             Drift mobility (cm /Vs)
                                                                                            2
Life time (ns)




                                                                                                                       0.15


                                                                                                                       0.10


                  2
                 10                                                                                                    0.05


                                                                                                                       0.00
                      40    60        80                              100   120      140                                   40         60          80   100     120   140
                                 Deposition rate (Å/s)                                                                                     Deposition rate (Å/s)


                      Figure 66. The lifetime as a function                                                             Figure 67. The drift mobility as a function
                               of deposition rate.                                                                                  of deposition rate.




                                                                                                 65
                             0.45
                             0.40
                                                                            (a)

Drift mobility (cm /Vs)
2
                             0.35
                             0.30
                             0.25
                             0.20                                       50Å/s
                             0.15                                       70Å/s
                                                                        90Å/s
                             0.10                                       110Å/s
                             0.05                                       130Å/s
                             0.00
                                       0   2000   4000    6000   8000   10000 12000
                                             Electric field (V/cm)
                                                                        50Å/s
                              10
                                   3                                    70Å/s
                                                                        90Å/s
                                                                        110Å/s
            Life time (ns)




                                                                        130Å/s




                                   2
                              10
                                                                            (b)
                                       0   2000   4000    6000   8000   10000 12000
                                             Electric field (V/cm)

Figure 68. Field dependence of the drift mobility (a) and the lifetime (b)
                   as a function of deposition rate.




                                                     66
                     0.16                                                                                     240
                                                                             (a)                                                                       (b)
                     0.15                                                                                     200




                                                                                         Range of LRPF (nm)
Depth of LRPF (eV)




                     0.14
                                                                                                              160
                     0.13
                                                                                                              120
                     0.12
                                                                                                               80
                     0.11

                     0.10                                                                                      40


                     0.09                                                                                       0
                         40    60        80                     100   120          140                           40     60        80     100     120         140
                                    Deposition rate (Å/s)                                                                    Deposition rate (Å/s)


                              Figure 69. The depth (a) and range (b) of the long range potential
                                         fluctuations as a function of deposition rate.



                                                            4



                                                            3
                                          N ∝ Vp /L (10 )
                                         -4




                                                            2
                                         2




                                                            1



                                                            0
                                                             40       60           80                         100     120        140
                                                                           Deposition rate (Å/s)

                                    Figure 70. The relative change in the density of charged defects as a
                                                        function of deposition rate.



                                                                                         67
Figure 70 shows the relative change in the charged defects as a function of deposition rate. It can
be seen that the charged defect density remains constant up to the deposition rate of ~110 Å/s, but
above, it begins to increase, similar to the case of the first series samples.


It is of great interest to ascertain what are the chemical or structural properties of the a-Si:H films
that vary as a function of the deposition rate. Brent Nelson has performed preliminary n&k
measurements on these new series samples. He found that on the stainless steel substrates, a
good fit on the analysis would occasionally be obtained, but an E04 value that was significantly
lower than that on the 1737 glass or other stainless steel substrates. When this would happen, the
thickness used in the fit consistently turned out to be thicker (lower E04, thicker sample by fit).
He and Qi Wang tried to sort out the n&k confusion, but do not have any good answers yet.
Perhaps, it is because the stainless steel substrates are not perfectly flat or a number of other
things. They have sent the samples to n&k Technologies to work with them in sorting this out,
but as of yet do not have any answers.




                                                  68
Light-induced decay measurements of transport properties of high deposition rate HWCVD a-
Si:H films


In phase II of this program, we reported the photomixing measurements of the first series of high
deposition rate HWCVD a-Si:H films in both the annealed the light-soaked states. The
deposition rates range from 32 to 191 Å/s. In the last quarterly report, we have investigated the
charge transport properties of the new series of HWCVD a-Si:H films in the annealed state. We
found that the results were similar to those for the first series prepared under non-optimized
growth conditions; the transport properties of the samples do not change monotonically with
increasing deposition rate, and peak in the neighborhood of ~70-90 Å/s, despite a monotonic
deterioration in the film performance was expected before. However, we observed there indeed
exist differences between two series of samples. Although the changes in the photoconductivity
with deposition rate are similar for two series, the drift mobility behaves in a different way
because the photoconductivity is proportional to the mobility-lifetime product.

In this report, we have characterized these new HWCVD a-Si:H samples in the light-soaked
states by using the photomixing technique, which enables us to determine the drift mobility,
lifetime and from the electric field dependence of these quantities, we have been able to
determine the range and the depth of the long-range potential fluctuations and from the latter two
quantities we can determine the relative changes of the charged defect density in the samples.
The light-soaking was performed at room temperature for about 3 hours by using a He-Ne laser
at the intensity of about 4 suns.

In order to investigate the influence of illumination on the transport properties of the samples as
a function of deposition rate, we performed systematical measurements on all five samples in
both the annealed and the light-induced states, whose deposition rates range from 50 to 130 Å/s.

The changes in the photoconductivity, drift mobility and lifetime for new series of samples with
light-soaking are similar to those for the first series; with light-soaking, the photoconductivity for
all the samples decreases, which is attributed to considerable decreases in the lifetime, while it is
interesting to observe that the drift mobility changes very small, even a little increase upon light-
soaking!




                                                 69
                                            4.5
            Photoconductivity (x10 S/cm)               State A
                                            4.0                                                     Figure 71. The light-induced
                                                       State B
                                            3.5                                                  changes in the photoconductivity as
        -4




                                                                                                    a function of deposition rate.
                                            3.0
                                            2.5
                                            2.0
                                            1.5
                                            1.0
                                            0.5
                                            0.0
                                              40        60        80      100        120   140
                                                             Deposition rate (Å/s)
                                           0.40
                                                      State A                                        Figure 72. The light-induced
                                           0.35       State B                                      changes in the drift mobility as a
Drift mobility (cm /Vs)




                                           0.30                                                       function of deposition rate.
2




                                           0.25

                                           0.20

                                           0.15

                                           0.10

                                           0.05

                                           0.00
                                               40       60        80      100        120   140
                                                             Deposition rate (Å/s)

                                                      State A
                                                      State B                                       Figure 73. The light-induced
                                              3
                                            10                                                   changes in the lifetime as a function
                                                                                                          of deposition rate.
                          Life time (ns)




                                              2
                                            10




                                                 40     60        80      100        120   140
                                                             Deposition rate (Å/s)

                                                                                 70
Figures 71-73 show the photoconductivity, drift mobility and lifetime, respectively, as a function
of deposition rate in both the annealed and the light-induced states. We can see that illumination
reduces the photoconductivity and the lifetime, while surprisingly increases the mobility for all
the samples, similar to the case of the first series of samples. However, it should be noted that the
changes in the mobility for the present samples as a function of deposition rate are opposite to
those for the first series (minimal µd vs maximal µd in the neighborhood of ~70-90 Å/s)!
Furthermore, the deposition rate dependence of these transport parameters in the light-soaked
state is similar to that in the annealed state. The light-induced decrease in the photoconductivity
is mainly due to the decrease in the lifetime since the mobility varies only by integer values,
while the lifetime varies by orders of magnitude.




                          0.55
                          0.50
                          0.45
Drift mobility (cm /Vs)




                          0.40
2




                          0.35
                          0.30
                          0.25
                          0.20
                                                                                    A        B   50Å/s
                          0.15                                                      A        B   70Å/s
                          0.10                                                      A        B   90Å/s
                                                                                    A        B   110Å/s
                          0.05                                                      A        B   130Å/s
                          0.00
                                      0       2000 4000 6000 8000 10000 12000
                                                Electric field (V/cm)
                          Figure 74. Effect of light-soaking on the electric field dependence of the drift
                                            mobility as a function of deposition rate.




                                                               71
                                                                       A         B   50Å/s
                       3
                  10                                                   A         B   70Å/s
                                                                       A         B   90Å/s
                                                                       A         B   110Å/s
                                                                       A         B   130Å/s
 Life time (ns)




                       2
                  10




                            0        2000 4000 6000 8000 10000 12000
                                      Electric field (V/cm)
                  Figure 75. Effect of light-soaking on the electric field dependence of the lifetime
                                           as a function of deposition rate.


Figures 74 and 75 show the electric field dependence of the drift mobility and lifetime,
respectively, of the samples as a function of deposition rate in both the annealed and the light-
induced states. It is seen that for lower fields, the light-induced changes in the drift mobility and
lifetime as a function of deposition rate are similar to the results shown in Figures 72 and 73.
Furthermore, it is clearly seen that the degree of the field dependence of the mobility and lifetime
become less after light-soaking, suggesting an enhanced effect of long-range potential
fluctuations.

From the field dependence of the drift mobility shown in Figure 74, we have determined the
depth and range of the long-range potential fluctuations. Figure 76 shows the influence of
illumination on the depth (a) and range (b) of potential fluctuations as a function of deposition
rate. We see that illumination reduces remarkably the range, as should be expected, but
meanwhile also reduce the depth a little which verify the light-induced increase in the mobility
observed above, in consistence with the results of the first series of samples.


                                                        72
                           0.16
                           0.15     State A                     (a)
                                    State B
                           0.14
    Depth of LRPF (eV)

                           0.13
                           0.12
                           0.11
                           0.10
                           0.09
                           0.08
                           0.07
                               40   60        80    100   120         140
                                     Deposition rate (Å/s)
                           240
                                    State A                     (b)
                           200      State B
      Range of LRPF (nm)




                           160

                           120

                            80

                            40

                             0
                              40    60        80    100   120         140
                                     Deposition rate (Å/s)

Figure 76. The light-induced changes in the depth (a) and range (b) of potential
                 fluctuations as a function of deposition rate.




                                               73
                                                                            2
Since the density of charged defects is proportional to VP / L, where VP and L represent the
depth and range of the long-range potential fluctuations, respectively, we can estimate the
relative change in the charged defect density by using the calculated values of VP and L. Figure
77 shows the light-induced relative change in the density of charged defects as a function of
deposition rate. Similarly, the charged defect density increases upon light-soaking for all
deposition rates, while there appears to be the least light-induced increase at the intermediate
deposition rates where the photoconductivity peaks (Fig.71). It should be noted that although the
density of charged defects increases upon light-soaking, the depth of the potential fluctuations
does not increase, but decreases. It appears that the light-induced increase in the charged defect
density results from the much more decrease in the range.




                    18              10
                                     9
                    16               8
                                     7
                    14
                         NLS/NAN (a.u.)




                                     6
                                     5
                                     4
 N ∝ V /L (x 10 )




                    12
-4




                                     3
                                     2
                    10               1
                                     0
                                     40   60   80 100 120           140
                     8                     Deposition rate (Å/s)
                                                                                State A
                     6
 2




                                                                                State B
p




                     4
                     2
                     0
                      40                  60            80            100       120       140
                                            Deposition rate (Å/s)
                    Figure 77. The light-induced relative change in the charged defect density as a
                                             function of deposition rate.



                                                               74
7.3     Characterization of high deposition rate HWCVD a-Si:H films deposited on stainless
        steel substrates by Driven Level Capacitance Spectroscopy and transient
        photocapacitance at the University of Oregon.

We were informed preliminary data from the University of Oregon as measured on the new
series of high deposition rate HWCVD a-Si:H samples by Driven Level Capacitance
Spectroscopy and transient photocapacitance. The results are shown below in Table 15.


Table 15. Characterization of high deposition rate HWCVD a-Si:H samples deposited on
          stainless steel substrates. (supplied by the University of Oregon).

                  Dep.
                            Thk.       EF             Eg       Eu        NDLCS         NPHCAP
Sample State      Rate
                            (µm)       (eV)           (eV)     (meV)     (1016 cm-3)   (1016 cm-3)
                  (Å/s)
           A                  1.45        0.69          1.72     53           <3            3.2
 L252                90
           B                              0.71                                <3            3.2
           A                  1.50        0.69          1.70     51          ~ 1.7          2.4
 L254               110
           B                              0.72                               ~2.2           3.6
           A                  0.77        0.69          1.70     49          1.25           6.2
 L255               130
           B                              0.72                                1.8           3.6


It is strange to observe that the defect density of L255 sample decreases after light-soaking as
determined by the Photocapacitance method. This is the sample giving the nicest DLCS profiles
and we can see that the DLCS defect density increases from State A (as-grown) to State B (120h
light-soaking at ~2W/cm2) as it should be. They suggested that the weird result in the
Photocapacitance is most likely due to the mobility-lifetime product of the holes. The bandtail of
the photocapacitance is "suppressed" and only the photocurrent measurement could give us an
idea by how much. It is more strongly suppressed for State A, thus making the defect density
look actually larger than it really is. The photocurrent measurements cannot be applied to an a-
Si:H film grown on a stainless steel substrate. Here is a reference which might clarify this issue
[22].

In addition, it is interesting to note that the Urbach energy Eu decreases monotonically with
increasing deposition rate, indicating more ordering network structure at higher deposition rates.
This is inconsistent with the trend with deposition rate, i.e., non-monotonic change, as measured
by the Constant Photocurrent Method (CPM) [21].

The present study indicates that the performance of a-Si:H films does not deteriorate
monotonically with increasing deposition rate as expected before. The transport properties (for
instance, the photoconductivity) of high deposition rate HWCVD a-Si:H films peak in the
neighborhood of ~70-90 Å/s, and good quality can be maintained at deposition rates up to 130
Å/s. The reason why the photoconductivity peaks in the neighborhood of ~70-90 Å/s may be due
to that this is where the pressure was increased at higher silane flows to increase the deposition
rate. So, there may have been significant depletion of the silane that changed things. The higher


                                                 75
pressure allows for more collisions before a dissociated molecule makes it from the filament to
the substrate. There is probably an optimal number of collisions to make good films.

From the results of two series of high deposition rate HWCVD a-Si:H samples, we have seen
some different behaviors in the transport properties. This might be attributed to different
microstructure in the films. Therefore, to further understand the transport properties of high
deposition rate a-Si:H samples, it is of great interest to perform detailed study of structural or
chemical properties as a function of deposition rate.




                                               76
7.4 The effect of deposition rate on the transport properties of HWCVD a-
    Si:H films with respect to the substrate temperature, deposition pressure
    and silane flow rate
In phase II of this program, we reported the photomixing transport measurements of the first set
of high deposition rate (32~191 Å/s) HWCVD a-Si:H films in both the annealed the light-soaked
states; we concentrated our attention on the effect of the deposition rate itself on the resultant
film properties. Since the high deposition rate (up to 1 µm/min.) was achieved by increasing
deposition pressure, silane flow rate, and decreasing filament-to-substrate distance, there exist
significant interactions among these parameters that require optimization to prepare films with
optimal quality for a desired deposition rate. Therefore, it is of great interest to investigate the
effect of the deposition rate on the resultant film properties with respect to the deposition
parameters, for instance, the deposition pressure, silane flow rate, substrate temperature, etc..
This would help us find out how to maintain high quality films at high deposition rates by means
of varying deposition parameters.


The preparation conditions of the first set of high deposition rate a-Si:H samples are shown in
Table 16. We had characterized these samples in the annealed state utilizing the photomixing
technique (see the 3rd Quarterly Report of Phase II). This technique enables us to determine the
drift mobility and lifetime separately; from the electric field dependence of these quantities, the
range and the depth of the long range potential fluctuations, and subsequently the relative
changes in the charged defect density in the films can be determined.


Figures 78-81, and Figure 82 show the deposition pressure dependence of the photomixing
transport parameters and hydrogen content, respectively, of the samples as a function of
deposition rate. It can be seen that the deposition rate is monotonically increased with increasing
deposition pressure; increasing pressure can increase deposition rate up to about 150 Å/s without
obviously deteriorating film quality although the film quality has a trend to decrease with
increasing pressure. Only too high pressure, beyond about 100 mTorr, will lead to significant
deterioration of film properties in the photoconductivity, drift mobility and lifetime as a result of
an increase in the concentration of charged defects, which lead to the long-range potential
fluctuations whose depth increases, while the range decreases. In addition, the hydrogen content
is reduced with increasing pressure (or deposition rate).




                                                 77
           Table 16. Preparation Conditions of High Deposition Rate HWCVD a-Si:H Samples.
            Sample Temperature Pressure             F (SiH4)     I (fil.) Thickness     Rate
               ID          (ºC)        (mTorr)       (sccm)        (A)       (Å)        (Å/s)
              L079         295            20           20           30      6644        31.64
              L077         295            70           20           30      8530        47.39
              L110         285            19           75           30     10514       58.41
              L139         315            50           50           29      8144        67.87
              L169         318            50           50           30      9235        76.96
              L104         350            50           50           30     10179       84.83
              L183         348            70           75           30     11000       109.78
              L178         317            70           75           30     11381       113.58
              L182         348            70           75           30     11502       114.79
              L190         326            50           75           30      9318       132.74
              L196         325            75           75           30      8538       142.30
              L094         385            90           75           30     13701       152.23
              L194         325            70           100          30     10662       177.70
              L087         295            120           75          30     22900       190.83




                                -7
                           10
Dark conductivity (S/cm)




                                                                                                 Photoconductivity (S/cm)




                                                                                                                             -3
                                                                                                                            10       58.4Å/s
                                                                                                                                                      109.8Å/s
                                                                           190.8Å/s
                                -8                                                                                           -4
                           10                                                                                               10                 132.7Å/s               152.2Å/s
                                                                                                                                                           142.3Å/s
                                                                152.2Å/s                                                     -5
                                                                                                                            10
                                     58.4Å/s
                                -9
                           10                                                                                                -6
                                               132.7Å/s                                                                     10

                                                                142.3Å/s                                                     -7
                                                                                                                            10                                           190.8Å/s
                                                     109.8Å/s
                            -10
                           10                                                   (a)                                                  (b)
                                                                                                                             -8
                                                                                                                            10
                                 0     20       40    60    80       100     120      140                                        0      20     40     60     80       100   120     140

                                      Deposition pressure (mTorr)                                                                      Deposition pressure (mTorr)


                                                      Figure 78. Deposition pressure dependence of the conductivity
                                                              of the samples as a function of deposition rate.




                                                                                            78
                          0.24                                                                                        300

                          0.20
Drift mobility (cm2/Vs)
                                                                                                                      250
                                    58.4Å/s                                                                                                          142.3Å/s
                          0.16                           109.8Å/s 152.2Å/s                                            200    58.4Å/s




                                                                                                   Lifetime (ns)
                          0.12                                                                                                          132.7Å/s
                                                 132.7Å/s                                                             150
                          0.08                                                                                                                                  152.2Å/s
                                                                                                                                                  109.8Å/s
                                                              142.3Å/s                                                100
                          0.04
                                                                                190.8Å/s
                          0.00                                                                                         50                                               190.8Å/s
                                   (a)                                                                                       (b)
                          -0.04                                                                                         0
                               0         20     40       60      80       100    120       140                           0      20      40     60       80      100      120       140

                                    Deposition pressure (mTorr)                                                               Deposition pressure (mTorr)

                                                  Figure 79. Deposition pressure dependence of the mobility (a)
                                                         and lifetime (b) as a function of deposition rate.




                          0.16                                                                                        120
                          0.15                                                    (a)                                                                                     (b)
                                                                                                                      100
                                                                                                 Range of LRPF (nm)




                                                               142.3Å/s
   Depth of LRPF (eV)




                          0.14                                                                                                                      142.3Å/s
                                                                                                                      80
                          0.13
                                              132.7Å/s
                          0.12                                                                                        60
                                                         109.8Å/s                                                                            109.8Å/s
                          0.11
                                                                                                                      40
                                                                     152.2Å/s
                          0.10                                                                                               58.4Å/s
                                   58.4Å/s                                                                            20                                     152.2Å/s
                          0.09                                                                                                         132.7Å/s

                          0.08                                                                                         0
                              0      20        40        60     80       100    120     140                             0      20      40     60      80       100      120     140

                                    Deposition pressure (mTorr)                                                               Deposition pressure (mTorr)


                                 Figure 80. Deposition pressure dependence of the depth (a) and range (b) of the
                                             potential fluctuations as a function of deposition rate.




                                                                                              79
                                          -3
              1.2x10                                                                                                                                                   12




                                                                                                                                             Hydrogen content (at.%)
                                          -3                                                                                                                           11
              1.0x10                                                                   152.2Å/s
                                                                                                                                                                       10         58.4Å/s
                                          -4
              8.0x10
N ∝ VP2 / L



                                                                                                                                                                           9
                                          -4
              6.0x10                                             132.7Å/s                                                                                                  8
                                                    58.4Å/s                                                                                                                                 132.7Å/s                152.2Å/s
                                          -4                                                                                                                               7
              4.0x10                                                           109.8Å/s
                                                                                                                                                                           6
                                          -4
              2.0x10
                                                                                  142.3Å/s                                                                                 5                       109.8Å/s 142.3Å/s
                                         0.0                                                                                                                               4
                                            0         20        40        60      80      100     120      140                                                              0          20     40       60      80      100     120    140

                                                     Deposition pressure (mTorr)                                                                                                   Deposition pressure (mTorr)

      Figure 81. Deposition pressure dependence of                                                                                                                             Figure 82. Deposition pressure
       the relative change in the density of charged                                                                                                                       dependence of the hydrogen content as a
         defects as a function of deposition rate.                                                                                                                               function of deposition rate.


     Figures 83-86 and Fig. 87 show the silane flow rate dependence of the photomixing transport
     parameters and hydrogen content, respectively, of the samples as a function of deposition rate.
     Similar to the case of increasing pressure, increasing silane flow rate also monotonically increase
     the deposition rate. However, the film quality does not change monotonically with varying silane
     flow rate; there exists an optimal flow rate of around 50 sccm, where the highest photoresponse,
     mobility and lifetime occur as a result of the lowest charged defect density, which is confirmed
     by the shallowest depth and the longest range of the potential fluctuations. This optimal flow rate
     corresponds to intermediate deposition rates between 70-90 Å/s. From Fig.87, it is seen that the
     hydrogen content does not seem to change with varying flow rate.
                                                                                                                  Photoconductivity (S/cm)
              Dark conductivity (S/cm)




                                                                                                177.7Å/s                                         10
                                                                                                                                                               -3
                                                                                                                                                                                              77.0Å/s


                                          10
                                               -9                         77.0Å/s
                                                       47.4Å/s                                                                                   10
                                                                                                                                                               -4
                                                                                                                                                                               47.4Å/s                      113.6Å/s

                                                                                     113.6Å/s
                                                                                                                                                               -5
                                                                                                                                                 10
                                                                                                                                                                                                                    177.7Å/s
                                                    (a)                                                                                                                     (b)
                                           -10                                                                                                                 -6
                                         10                                                                                                      10
                                                0          20        40         60        80      100      120                                                         0          20        40         60      80      100      120

                                                                Silane flow (sccm)                                                                                                      Silane flow (sccm)
                                                                     Figure 83. Silane flow rate dependence of the conductivity
                                                                           of the samples as a function of deposition rate.
                                                                                                                 80
                            0.24                                                                               1200
                                                                                 (a)                                                                                             (b)
Drift mobility (cm2/Vs)


                            0.20                                                                               1000
                                                                                                                                                       77.0Å/s
                            0.16
                                                  77.0Å/s                                                             800




                                                                                               Lifetime (ns)
                                                                 113.6Å/s
                            0.12                                                                                      600
                                                                                                                      400
                            0.08                                                                                                                                 113.6Å/s
                                     47.4Å/s
                                                                            177.7Å/s                                  200            47.4Å/s
                            0.04                                                                                                                                              177.7Å/s
                                                                                                                               0
                            0.00
                                0         20     40     60          80       100       120                                       0     20         40       60       80        100      120

                                               Silane flow (sccm)                                                                               Silane flow (sccm)


                                                   Figure 84. Silane flow rate dependence of the mobility (a)
                                                       and lifetime (b) as a function of deposition rate.




                            0.16                                                                                               120

                                                                                                                               100
                                                                                                          Range of LRPF (nm)
       Depth of LRPF (eV)




                            0.14                                              177.7Å/s
                                                                                                                                                       77.0Å/s
                                      47.4Å/s                                                                                  80
                                                                                                                                                                   113.6Å/s
                            0.12                                                                                               60     47.4Å/s
                                                                   113.6Å/s
                                                   77.0Å/s                                                                     40
                            0.10                                                                                                                                              177.7Å/s
                                                                                                                               20
                                    (a)                                                                                              (b)
                            0.08                                                                                                0
                                0         20      40        60       80       100        120                                     0         20      40        60       80        100      120

                                               Silane flow (sccm)                                                                                Silane flow (sccm)


                                                Figure 85. Silane flow rate dependence of the depth (a) and range (b)
                                                    of the potential fluctuations as a function of deposition rate.




                                                                                                   81
                   -3
              1.6x10                                                                                  8




                                                                            Hydrogen content (at.%)
                   -3
              1.2x10                                       177.7Å/s                                   7
                                                                                                                                113.6Å/s
N ∝ VP2 / L




                   -4
              8.0x10                                                                                  6              77.0Å/s

                           47.4Å/s
                                                                                                                                           177.7Å/s
                   -4
              4.0x10                                113.6Å/s                                               47.4Å/s
                                                                                                      5
                                          77.0Å/s

                  0.0
                                                                                                      4
                       0     20      40       60      80       100    120                              0    20       40    60      80        100      120

                                  Silane flow (sccm)                                                             Silane flow (sccm)


                  Figure 86. Silane flow rate dependence of                          Figure 87. Silane flow rate dependence of
                     the relative change in the density of                             the hydrogen content dependence of the
                                                                                       Fig.87 Silane flow rate as a function of
                  charged defects as a function of deposition                          hydrogen content as a function of
                                                                                                   deposition rate.
                                     rate.                                             deposition rate.




            Figures 88-91 and Figure 92 show the substrate temperature dependence of the photomixing
        transport parameters and hydrogen content, respectively, of the samples as a function of
        deposition rate. It is seen that as anticipated, the hydrogen content is reduced with increasing
        substrate temperature, but the substrate temperature does not have a significant influence on the
        deposition rate. However, it is worthy of note that at higher deposition rates, high quality films
        are grown at higher substrate temperature, suggesting that in order to produce high quality films
        at high rates, higher substrate temperatures are required.

           From the above results, we see that low pressure, medium silane flow rate, and higher
        substrate temperature are generally required to maintain high quality a-Si:H films at high
        deposition rates. While, we should realize that due to significant interactions among these
        parameters, optimization is required to prepare films with optimal quality for a desired
        deposition rate. Therefore, further work is needed to perform well-defined studies of the
        deposition rate dependence of film properties, both electronic and structural, under otherwise
        similar circumstances, i.e., without significant change in film grown precursors and discharge
        conditions.




                                                                      82
                                                          -7
                                                     10                                                                                                          -3
                                                                                                                                                                10




                                                                                                                                     Photoconductivity (S/cm)
                                                                                                                  (a)                                                                        132.7Å/s
                          Dark conductivity (S/cm)




                                                                                                                                                                 -4    58.4Å/s
                                                                          190.8Å/s                                                                              10                113.6Å/s           109.8Å/s
                                                          -8
                                                     10                                                                                                                                                          152.2Å/s
                                                                                                                                                                 -5
                                                                                                                                                                10

                                                               58.4Å/s                                                                                           -6
                                                          -9                                                                                                    10
                                                     10                                                     152.2Å/s
                                                                                      132.7Å/s
                                                                                                                                                                 -7
                                                                                                                                                                10             190.8Å/s
                                                                         113.6Å/s                 109.8Å/s                                                                                                               (b)
                                                      -10                                                                                                        -8
                                                     10                                                                                                         10
                                                          260     280      300       320   340      360         380     400                                       260    280      300     320      340    360       380        400
                                                                                                            o                                                                                                    o
                                                                  Substrate temperature ( C)                                                                              Substrate temperature ( C)


                                                                  Figure 88. Substrate temperature dependence of the conductivity
                                                                           of the samples as a function of deposition rate.



                                      0.24                                                                                                                  240
Drift mobility (cm /Vs)




                                      0.20                                                                152.2Å/s
                                                                                                                                                            200
2




                                      0.16                                                                                                                                              113.6Å/s
                                                                                                 109.8Å/s
                                                                                                                              Lifetime (ns)




                                                                           113.6Å/s                                                                                                                      109.8Å/s
                                      0.12                                                                                                                  160         58.4Å/s
                                                                58.4Å/s
                                                                                     132.7Å/s
                                      0.08                                                                                                                  120                              132.7Å/s

                                      0.04                                                                                                                                                                           152.2Å/s
                                                                   190.8Å/s                                                                                                      190.8Å/s
                                                                                                                                                                80
                                      0.00
                                                                                                                 (a)                                                                                                      (b)
                                 -0.04                                                                                                                          40
                                     260                        280      300     320       340     360       380        400                                      260     280      300     320      340     360        380       400
                                                                                                            o                                                                                                        o
                                                                Substrate temperature ( C)                                                                               Substrate temperature ( C)



                                                                         Figure 89. Substrate temperature dependence of the mobility (a)
                                                                                 and lifetime (b) as a function of deposition rate.




                                                                                                                              83
                             0.13                                                                                         60
                                                     132.7Å/s                   (a)                                                                                                 (b)
                                                                                                                          50




                                                                                             Range of LRPF (nm)
                             0.12       113.6Å/s
        Depth of LRPF (eV)




                                                                         152.2Å/s                                                             113.6Å/s              109.8Å/s
                                                                                                                          40
                             0.11
                                                            109.8Å/s
                                                                                                                          30
                             0.10
                                      58.4Å/s                                                                                                            132.7Å/s
                                                                                                                          20
                                                                                                                                                                               152.2Å/s
                             0.09                                                                                                        58.4Å/s
                                                                                                                          10
                             0.08
                                260   280   300    320     340     360      380       400                                            280     300     320    340       360      380        400
                                                                            o                                                                                                  o
                                      Substrate temperature ( C)                                                                         Substrate temperature ( C)


                                       Figure 90. Substrate temperature dependence of the depth (a) and range (b)
                                               of the potential fluctuations as a function of deposition rate.




                                -3
                 1.2x10                                                                                                             12
                                                                                                                                    11
                                                                                                          Hydrogen content (at.%)




                                -4                                       152.2Å/s                                                   10     58.4Å/s
                 9.0x10
                                                                                                                                    9
N ∝ VP2 / L




                                -4
                 6.0x10                             132.7Å/s                                                                        8
                                      58.4Å/s                                                                                                              132.7Å/s                152.2Å/s
                                                                                                                                    7
                                                                 109.8Å/s
                                -4
                                                                                                                                    6                                 109.8Å/s
                 3.0x10                                                                                                                        113.6Å/s
                                                113.6Å/s                                                                            5
                              0.0                                                                                                   4
                                260 280 300 320 340 360 380 400                                                                     260 280 300 320 340 360 380 400
                                                                            o                                                                                                      o
                                      Substrate temperature ( C)                                                                           Substrate temperature ( C)

                                 Figure 91. Substrate temperature                                                                       Figure 92. Substrate temperature
                              dependence of the relative change in the                                                               dependence of the hydrogen content as a
                             density of charged defects as a function of                                                                   function of deposition rate.
                                           deposition rate.



                                                                                        84
8. Measurements of amorphous (Si,Ge) alloys

8.1   Measurements of homogenous a-SiGe alloy samples produced at NREL

The a-SiGe:H sub-team at NREL is presently employing the hot-wire technique to produce a-
SiGe:H alloys which are to be optimized with respect to the transport parameters within two
bandgap regimes: 1.40-1.45 eV to 1.60-1.65 eV. We have previously determined the transport
parameters of a series a-SiGe:H alloys supplied by Brent Nelson where he explored the
parameter space involving the number of collisions a reactive species makes before impinging on
the substrate surface. These measurements involved a range of preparation parameters which
resulted in inhomogeneity of composition. We now report on a series of samples that are
homogeneous in alloy composition supplied by Nelson. The photoconductivity, drift mobility,
and lifetime were studied systematically in the annealed and light-soaked states employing the
photomixing technique using the longitudinal modes of a HeNe laser. All the samples were
initially annealed at 1500C for 2 hours to restore the samples to the annealed state. The
characteristics of the samples are listed in Table 17.


Table 17. The characteristics of a-SiGe:H and a-Si:H.

             Sample ID     Substrate Temp.     GeH4 (%)     Thickness (µm)
                                  0
                                    C
              HGe118             400                 0           1.093
              HGe114             475                 3           1.447
              HGe121             365                 3           1.858
              HGe113             475                 8           1.482
              HGe100             500                17           2.055


Figure 93 shows the normalized photoconductivity, mobility, and lifetime for a-Si:H prepared by
the same hot-wire technique as a function of illumination time. During illumination, both the
drift mobility and lifetime decrease. This is consistent with previous results we have obtained
indicating that both neutral and charged defects are created by light-soaking.

Figures 94-97 show the normalized photoconductivity, mobility, and lifetime in a-SiGe:H alloys
prepared with different GeH4 gas flow ratios as a function of illumination time. It can be seen
from these results that both the mobility and lifetime are lowered by light-soaking which is
similar to a-Si:H; however it should be noted that the rate of decrease of these quantities varies
as a function of the sample composition. We have also measured the electric field dependence of
the mobility and thus determined the range and the depth of the long-range potential fluctuations.
The results for the annealed state are shown in Table 18 while the results for the light-soaked
state are shown in Table 19.




                                               85
                                  1.2



Normalized decay ratio (a.u)
                                  1.0

                                  0.8

                                  0.6      HGe118
                                            Drift mobility
                                  0.4       Lifetime
                                            Photoconductivity
                                  0.2
                                   0.01      0.1     1       10    100
                                              Illumination time (min)

 Figure 93. The normalized photoconductivity, mobility, and lifetime
             in a-Si:H as a function of illumination time.
                                  1.2
  Normalized decay ratio (a.u.)




                                  1.0

                                  0.8

                                  0.6       HGe114
                                              Drift mobility
                                  0.4         Lifetimety
                                              Photoconductivity
                                  0.2
                                    0.01     0.1     1      10      100
                                              Illumination time (min)

 Figure 94. The normalized photoconductivity, mobility, and lifetime
  as a function of illumination time. The GeH4 gas flow ratio is 3%.




                                                     86
                                          1.2

Normalized decay ratio (a.u.)             1.0

                                          0.8

                                          0.6                     HGe121
                                                                   Drift mobility
                                          0.4                      Lifetime
                                                                   Photoconductivity
                                          0.2
                                            0.01                  0.1        1         10         100
                                                                    Illumination time (min)
                                                 Figure 95. The normalized photoconductivity, mobility, and lifetime
                                                  as a function of illumination time. The GeH4 gas flow ratio is 3%.


                                                         1.2
                         Normalized decay ratio (a.u.)




                                                         1.0

                                                         0.8

                                                         0.6        HGe113
                                                                    Drift mobility
                                                         0.4        Lifetime
                                                                    Photoconductivity
                                                         0.2
                                                           0.01    0.1      1         10      100
                                                                     Illumination time (min)
                                                 Figure 96. The normalized photoconductivity, mobility, and lifetime
                                                  as a function of illumination time. The GeH4 gas flow ratio is 8%.



                                                                                 87
       Normalized decay ratio (a.u.)   1.2

                                       1.0

                                       0.8

                                       0.6
                                                    HGe100
                                       0.4          Drift mobility
                                                    Lifetime
                                                    Photoconductivity
                                       0.2
                                                   1               10               100
                                                       Illumination time (min)

                                       Figure 97. The normalized photoconductivity, mobility, and lifetime
                                       as a function of illumination time. The GeH4 gas flow ratio is 17%.




Table 18. The photoconductivity, drift mobility, lifetime, and range and depth of long range
          potential fluctuations in annealed state for a-SiGe:H alloys prepared with different
          GeH4 gas flow ratio.

                                               HGe100      HGe113       HGe121      HGe114       HGe118
        GeH4                                      17%         8%           3%          3%           0%
    σph (Ω-1cm-1)                              0.31×10-4   3.72×10-4    1.39×10-4   3.40×10-4    5.91×10-4
        τ (ns)                                   49.87      179.00        62.42       86.00        71.28
    µd (cm2/Vs)                                  0.067       0.160        0.216       0.303        0.473
       L (nm)                                    25.61       34.46        29.05       38.24        41.24
      Vp (eV)                                    0.143       0.117        0.109       0.103        0.092




                                                                        88
Table 19. The photoconductivity, drift mobility, lifetime, and range and depth of long range
          potential fluctuations in light-soaked state for a-SiGe:H alloys prepared with
          different GeH4 gas flow ratio.

                                        HGe100              HGe113            HGe121          HGe114      HGe118
         GeH4                              17%                 8%                3%              3%          0%
     σph (Ω-1cm-1)                      0.27×10-4           1.46×10-4         0.67×10-4       1.53×10-4   2.21×10-4
         τ (ns)                           41.96               87.72             32.60           49.33       30.67
     µd (cm2/Vs)                          0.065               0.133             0.194           0.242       0.425
        L (nm)                            57.40               29.00             20.62           21.29       26.79
       Vp (eV)                            0.158               0.123             0.109           0.104       0.093


In the annealed state, the photoconductivity, drift mobility, lifetime and the range of the long-
range potential fluctuations in a-SiGe:H decrease, while the depth of the long-range potential
fluctuations increases with increasing GeH4 gas flow ratio. The number of charge states can be
estimated using the range and depth of the long-range potential fluctuations due to charge states;
some of the charge states can be due to the effect of alloying whereby the neutral homopolar
bonds of Ge-Ge and Si-Si get modified in the case of a Ge-Si bonds resulting in a slight ionic
character.

The surface morphology of these samples were determined by AFM in cooperation with Dr.
David Braunstein formerly of Parks Scientific and presently of IBM. Figure 98 shows the RMS
surface roughness vs. the GeH4 gas flow rate. The transport parameters as a function of the GeH4
gas flow rate are shown in Figure 99. Figures 100-104 show the images of the surface scans and
the respective surface roughnesses. The surface roughness should not be interpreted as grain size.
To arrive at the grain size it is necessary to perform a Fourier analysis of the image which will be
reported a later date.


                                   28
                                                                                     o
                                   26                                            500 C
              RMS Roughness (nm)




                                                                o
                                             o              475 C
                                   24   400 C


                                   22
                                                                                          2
                                                            Scanned area: (0.5µm)
                                   20               o
                                                 475 C
                                   18
                                         0              5           10         15         20
                                                   GeH4 gas ratio (%)
                                         Figure 98. RMS roughness vs. GeH4 gas ratio.


                                                                         89
                                2.5
 Photoconductivity [10 Ω cm ]
-1




                                2.0
-1
-4




                                1.5

                                1.0

                                0.5

                                0.0
                                      0        5      10      15    20
                                             GeH4 gas flow rate
                                0.5                                                          90


                                0.4                                                          80
Mobility [cm / Vs]




                                                                             Lifetime [ns]
                                                                                             70
                                0.3
2




                                                                                             60

                                0.2                                                          50


                                0.1                                                          40


                                                                                             30
                                0.0
                                      0        5      10     15     20                            0        5      10     15    20

                                          GeH4 gas flow ratio (%)                                       GeH4 gas flow rate

                                 60                                                    0.16


                                 50
            Range [nm]




                                                                                       0.14
                                                                         Depth [eV]




                                 40
                                                                                       0.12
                                 30

                                                                                       0.10
                                 20

                                      0        5      10      15    20                 0.08
                                                                                                  0        5      10     15    20
                                          GeH4 gas flow rate (%)
                                                                                                      GeH4 gas flow rate (%)

                                      Figure 99. Photoconductivity, mobility, lifetime, LRPF range and depth
                                             vs. germanium hydrogen gas flow ratio (annealed state).




                                                                                  90
Figures 100-102. The AFM scanning images and surface height
      distribution of the samples HGe 121, 113, and 100.




                              91
Figures 103-104. The AFM scanning images and surface height
      distribution of the samples HGe 114 and HGe 118.




                              92
8.2   Charge transport properties of PECVD a-SiGe:H films produced by BP Solar

We have employed the photomixing technique to study the charge transport properties of
HWCVD a-SiGe:H alloys as a function of alloy composition [23]. Evidence for the presence of
long-range potential fluctuations in a-SiGe:H was revealed from the measurements of electric
field dependence of the drift mobility, and the effect of the long-range potential fluctuations is
enhanced by the addition of Ge to the alloy system that results in the deterioration of the opto-
electronic properties of a-SiGe:H. It was found that at a composition of ~10% Ge in Si, the
photoresponse begins to decrease monotonically with increasing Ge content due to the decreases
in both the drift mobility and the lifetime as a result of an increase in the concentration of
charged defects, which lead to the long-range potential fluctuations whose depth increases, while
the range decreases. It is of interest to study the influence of the incorporated Ge on a-SiGe:H
alloys prepared by other techniques.

We have received a series of a-SiGe:H samples prepared by the PECVD technique from BP
Solar for photomixing transport measurements. These samples were deposited on Corning glass
substrates with varying Ge content (0-40%). Table 20 shows the Ge content and film thickness
for the samples being studied. Detailed deposition conditions of the samples have not been
disclosed at this time.

Table 20. The Ge content and film thickness for the samples produced by BP Solar.

      Sample ID          Ge content (%)              Thickness (µm)
      A0270-1                   40                         0.92
       A0270-2                  20                         0.87
      A0271-1                   10                         0.70
      A0271-2                   5                          0.74
      A0272-1                   0                          0.70


Figures 105-108 show the photoconductivity, the drift mobility and lifetime, the depth and range
of the potential fluctuations, and the relative changes in the charged defect density, respectively,
of the samples as a function of Ge content. It is seen that the results for PECVD a-SiGe:H
samples are basically similar to those for HWCVD materials, i.e., the photoconductivity is
reduced monotonically with increasing Ge content due to a decrease in the mobility as a result of
an increase in the concentration of charged defects, which lead to the long-range potential
fluctuations whose depth increases, while the range decreases. While, the lifetime shows little
change, and surprisingly even a little increase with increasing Ge content, in contrast to the case
for HWCVD materials where the lifetime decreases with increasing Ge content. Moreover, from
the data shown in the figures, it appears that a little more Ge, up to ∼20%, is allowable to be
added to the alloy system before the film transport properties are deteriorated significantly in the
case of PECVD materials, consistent with the previous results [24].




                                                93
                                                                        10




                                         Photoconductivity (x10 S/cm)
                                                                         8




                                         -4
                                                                         6


                                                                         4


                                                                         2


                                                                         0
                                                                             0     10         20        30                    40       50
                                                                                       Ge content (%)


                  Figure 105. The photoconductivity of the samples as a function of Ge content.




                          0.40                                                                                          500
                                                                                                                        400
                          0.35                                                            (a)                                                                 (b)
                                                                                                                        300
Drift mobility (cm /Vs)




                          0.30
                                                                                                                        200
2




                                                                                                        Lifetime (ns)




                          0.25

                          0.20                                                                                          100
                                                                                                                         90
                                                                                                                         80
                          0.15                                                                                           70
                                                                                                                         60
                                                                                                                         50
                          0.10
                                                                                                                         40
                          0.05                                                                                          30

                          0.00                                                                                          20
                                 0   5                                  10   15   20     25        30                              0    5    10   15   20    25     30

                                                 Ge content (%)                                                                             Ge content (%)


                           Figure 106. The mobility (a) and lifetime (b) of the samples as a function of Ge




                                                                                                94
                     0.14                                                                                         25


                     0.13




                                                                                             Range of LRPF (nm)
Depth of LRPF (eV)




                     0.12                                                                                         20


                     0.11


                     0.10                                                                                         15


                     0.09
                                                                                 (a)                                                                      (b)
                     0.08                                                                                         10
                              0     5       10                 15       20   25        30                              0    5        10        15   20   25     30

                                        Ge content (%)                                                                          Ge content (%)
                            Figure 107. The depth (a) and range (b) of potential fluctuations as a function of Ge
                                                                 content.

                                                              10



                                                              8
                                         N ∝ VP / L (x 10 )
                                        -4




                                                              6
                                        2




                                                              4



                                                              2
                                                                    0        5         10              15              20       25        30
                                                                                  Ge content (%)

                                        Figure 108. The relative change in the density of charge defects
                                                         as a function of Ge content.




                                                                                            95
8.3   Electrical and optical properties of high quality low bandgap amorphous (Ge, Si)
      alloys prepared by ECR plasma deposition


We have employed the photomixing technique to study the charge transport properties of a-
SiGe:H alloys growth by both HWCVD and PECVD as a function of alloy composition. It was
found that the quality of both HWCVD and PECVD materials and the device performance
degrade significantly with alloys having >40% Ge content being rather poor in quality so that the
transport parameters at the Ge end could not be determined by the photomixing technique. In
general, this deterioration in quality is ascribed to poorer microstructure of the materials with
high Ge content.

It is noteworthy that the monotonic increase in the depth and the monotonic decrease in the range
of the potential fluctuations with increasing Ge content up to 40% in this alloy system suggest
that compositional disorder may also play a role in the long-range potential fluctuations. If this is
the case, at a certain critical Ge content, there should exist the largest degree of the
compositional disorder in the alloy system that will results in the worst film quality, as may be
the case in crystalline SixGe1-x alloys [25]; beyond this, the film quality would increase again.
Therefore, it is of great interest to pursue a suitable technique to produce good quality a-GeSi:H
alloys at the Ge end, even pure a-Ge:H films, and to perform optoelectronic and structural
investigations of the alloys and related devices.

Recently, we have received a series of high quality low bandgap a-(Ge, Si):H alloys at the Ge
end from the Dalal group. These a-(Ge, Si):H samples were prepared using low pressure,
reactive ECR plasma deposition with high H dilution and ppm B-doping. Incorporating these
high quality materials into devices leads to much lower gap a-(Ge, Si) solar cells (down to ∼1 eV
in a-Ge:H) with acceptable performance [26]. Here, we report on the charge transport and optical
properties of these low bandgap materials as a function of alloy composition investigated by
employing the microwave photomixing technique and optical absorption spectroscopy. Table 21
shows the growth conditions and some measured properties of ECR a-(Ge, Si):H samples
supplied by the Dalal group.

Figure 109 shows the optical absorption spectra of the samples measured with a Lambda 9
spectrophotometer, where the Si content and the Tauc gap Eg are also given. It is seen that with
increasing Si content, the absorption edge shifts to higher energy, indicating an increase in Eg,
from ∼1 eV for a-Ge:H up to 1.34 eV at SiH4/(SiH4+GeH4) = 57%. At the other end of the
composition range (the Si end), the absorption edge shifts to lower energy when Ge is added to
Si, as observed previously.

Figure 110 shows the sub-gap absorption spectra of some of the samples obtained using a dual
beam quantum efficiency apparatus, where the Si content and the Urbach energy (Eo) of the
valence band-tail are also given. The Urbach energy Eo was obtained from the exponential part
of the absorption spectra. The Urbach tail is known to be due to bonding and thermal disorder in
the material, and may be used to study the disorder in the network structure [27,28]. We can see
that for pure a-Ge:H, there is strong sub-gap absorption due to high density of gap defect states


                                                 96
  below the Fermi level, but the Urbach energy Eo is low, comparable to that for a-Si:H. With
  increasing Si content, the sub-gap absorption (or the gap-state density) is reduced significantly,
  while the Urbach energy Eo is increased considerably, up to more than 90 meV at 44% Si in Ge,
  suggesting an increase in the compositional disorder.


  Table 21. Growth conditions and properties of ECR a-GeSi:H/a-Ge:H samples.

                                               SiH4        GeH4    ppb TMB
Sample   Temp.    Press.    Power     H2                                     Thk.      σd          σph
                                              (100%)      (10%)    (10ppm)
ID        (ºC)   (mTorr)     (W)    (sccm)                                   (µm)    (S/cm)      (S/cm)
                                              (sccm)      (sccm)    (sccm)
2/4392    350       10       150      42         0         9.6        4      0.70   7.00×10-6   8.80×10-6
2/4393    350       5.6      150      42        1.9        14.4       4      0.87   1.70×10-9   1.87×10-7

2/4394    350       5.2      150      42       1.14        14.4       4      0.98   2.00×10-8   1.28×10-6

2/4395    350       5        150      42       1.52        12.8       4      0.82                 
2/4396    350       5.8      150      42       0.95        16         4      1.62   2.30×10-8   1.09×10-6

2/4397    300       5.9      150      42         0         16         4      1.07   7.90×10-5   9.60×10-5

2/4399    300       10       150      42         0         9.6        4      1.30   1.44×10-4   1.60×10-4



  Figure 111 shows the photoconductivity (proportional to mobility-lifetime product µdτ) of the
  samples as a function of the Tauc gap and therefore alloy composition. It is significant to see
  from Fig.111 and Table 21 that an abrupt decrease (more than one order of magnitude) in µdτ
  from the a-Ge:H endpoint as Si is added to the system sets in the neighborhood of 30%
  SiH4/(SiH4+GeH4), and then continues to decrease with increasing Si content up to 57%. In
  contrast to the a-SiGe:H films at the Ge end prepared by HWCVD or PECVD, there is no
  photoresponse observed when GeH4/(SiH4+GeH4) is over 40% (or 60% Si content) so that the
  photomixing technique could not determine their transport parameters, indicating very poor
  performance. The present results indicate that the transport properties of a-GeSi:H alloys at the
  Ge end produced by reactive ECR plasma deposition have been improved significantly. The
  much improved film quality allows us to separately determine the drift mobility and lifetime;
  from the measurements of electric field dependence of the drift mobility, the depth and range of
  the potential fluctuations, and subsequently the charged defect density can be determined as a
  function of alloy composition.

  Figure 112 shows the mobility (a) and lifetime (b) as a function of alloy composition. We see
  that both µd and τ, particularly µd, decrease abruptly at ∼30% SiH4/(SiH4+GeH4); with further
  increasing Si content, µd changes very little, but τ continues to decrease. It is thus evident that
  the Si-induced degraded photoresponse at the Ge end can be attributed to the decreases in both
  µd and τ, similar to the case of the Ge-induced decay at the Si end.

  Figure 113 shows the electric field dependence of the mobility of the samples as a function of
  alloy composition. The solid curves shown in the figure are fits of the data to equation (1). Due


                                                     97
to the presence of long-range potential fluctuations, an obvious consequence is that the µd should
increase with an applied field, because the external field offsets the internal field and reduces the
magnitude of the potential barrier. Furthermore, it is clearly seen that with increasing Si content
in the alloys, the absolute values of the mobility are reduced significantly in the whole electric
field range measured; moreover, with the addition of Si to the alloys, the degree of the field
dependence of the mobility becomes less, suggesting an enhanced effect of long-range potential
fluctuations, which will be confirmed below by the determination of the depth and range of
potential fluctuations as a function of alloy composition.

From the electric field dependence of the mobility shown in Fig. 113, the depth and range of
long-range potential fluctuations can be inferred (Fig.114 (a) and (b)), and consequently, the
relative change in the density of charged defects can be estimated (Fig.115). It is seen that the
depth begins to increase, the range decreases and the charged defect density increases beyond
∼30% SiH4/(SiH4+GeH4), indicating that the effect of long-range potential fluctuations is
enhanced as Si is added to the alloy system.

In summary, we have investigated the charge transport and optical properties of high quality low
bandgap a-(Ge, Si):H alloys at the Ge end as a function of alloy composition by employing the
microwave photomixing technique and optical absorption spectroscopy. It was found that at
SiH4/(SiH4+GeH4) ~30%, the photoresponse begins to decrease rapidly with increasing Si
content due to the decreases in the mobility and lifetime, and meanwhile, both the charged defect
density and the Urbach energy increase significantly. The latter indicates an increase in the
compositional disorder. It is the potential fluctuations whose effect can be also enhanced by
incorporating Si to the alloy system that result in the deterioration of the opto-electronic
properties of a-(Ge, Si):H alloys, similar to the case of the incorporation of Ge at the Si end. This
enhanced effect accompanies with an increase in the depth, and a decrease in the range of
potential fluctuations, leading to a decrease in the mobility, and consequently in the
photoconductivity. The increased charged scattering centers and compositional disorder upon
adding Si or Ge to the alloys observed play an important role in the potential fluctuations.

Further work is needed to perform detailed study of high quality low bandgap a-(Ge, Si):H alloys
across the entire range of Ge content (0%-100%) in order to find out the effect of the
compositional disorder on the film properties.




                                                 98
             5
          10




             4
          10
α (cm )
-1




                                                      2/4392 (0% Si, Eg=1.09eV)
                                                      2/4393 (57% Si, Eg=1.34eV)
                                                      2/4394 (44% Si, Eg=1.31eV)
             3
          10                                          2/4395 (54% Si, Eg=1.21eV)
                                                      2/4396 (37% Si, Eg=1.25eV)
                                                      2/4397 (0% Si, Eg=1.06eV)
                                                      2/4399 (0% Si, Eg=1.02eV)

                    1.0          1.2           1.4          1.6           1.8          2.0
                                              hν (eV)


     Figure 109. The optical absorption spectra of the samples. The Si content and Tauc gapEg




                                               99
               4
          10

               3
          10

               2
          10
α (cm )
-1




               1
          10

                                                                       2/4392 (0% Si, Eo=31.2 meV)
               0
          10                                                           2/4394 (44% Si, Eo=91.8 meV)
                                                                       2/4396 (37% Si, Eo=48.3 meV)
                                                                       2/4399 (0% Si, Eo=46.1 meV)
               -1
          10
                                   0.8                   1.0    1.2         1.4         1.6     1.8
                                                                 hν (eV)
  Figure 110. The sub-gap absorption spectra of some of the samples. The Si content and
              the Urbach energy (Eo) of the valence band-tail are also given.
                    Photoconductivity (S/cm)




                                                    -4
                                               10




                                                    -5
                                               10

                                                         1 .0   1 .1       1 .2        1 .3   1 .4
                                                                T a u c g a p (e V )

      Figure 111. The photoconductivity of the samples as a function of the Tauc gap.



                                                                   100
Drift mobility (cm2/Vs)       0.03

                                                                                                                                                                               3
                                                                                                                                                                         10
                              0.02




                                                                                                                                              Lifetime (ns)
                              0.01



                              0.00
                                        (a)                                                                                                                                          (b)
                                                                                                                                                                               2
                                                                                                                                                                         10
                                         1.0                                1.1        1.2          1.3                              1.4                                              1.0          1.1                              1.2            1.3          1.4
                                                                            Tauc gap (eV)                                                                                                          Tauc gap (eV)

    Figure 112. The mobility (a) and lifetime (b) of the samples as a function of the Tauc
                                            gap.


                                                                                                                                                                                                                                   0.10
                              0.020     LRPF Depth: 0.28 eV                                                                   0.16     LRPF Depth: 0.24 eV                                                                                    LRPF Depth: 0.27 eV
                                        LRPF Range: 1.8 µm                                                                             LRPF Range: 10 µm                                                                           0.08       LRPF Range: 10 µm
    Drift mobility (cm2/Vs)




                                                                                                    Drift mobility (cm2/Vs)




                                                                                                                                                                                                         Drift mobility (cm2/Vs)
                              0.015                                                                                           0.12
                                                                                                                                                                                                                                   0.06
                                       57% Si, Eg=1.34eV
                              0.010                                                                                           0.08                                                                                                 0.04

                                                                                                                                                                                                                                   0.02
                              0.005                                                                                           0.04

                                                                                                                                                                                                                                   0.00
                              0.000                                                                                           0.00                                            44% Si, Eg=1.31eV                                                              54% Si, Eg=1.21eV
                                                                                                                                                                                                                                   -0.02
                                      3000 4000 5000 6000 7000 8000                                                                   5000             6000                          7000      8000                                         4000    5000      6000    7000   8000

                                                           Electric field (V/cm)                                                             Electric field (V/cm)                                                                                 Electric field (V/cm)

                                                                     0.16                                                                                                     0.08
                                                                              LRPF Depth: 0.26 eV                                                                                      LRPF Depth: 0.24 eV
                                                                              LRPF Range: 10 µm                                                                                        LRPF Range: 10 µm
                                                                     0.12                                                                                                     0.06
                                           Drift mobility (cm2/Vs)




                                                                                                                                                    Drift mobility (cm2/Vs)




                                                                                                                                                                                      0% Si, Eg=1.02eV
                                                                     0.08                                                                                                     0.04


                                                                     0.04                                                                                                     0.02


                                                                     0.00                                                                                                     0.00
                                                                                              37% Si, Eg=1.25eV
                                                                       2000 3000 4000 5000 6000 7000 8000                                                                       1000        1500   2000                   2500             3000    3500
                                                                                  Electric field (V/cm)                                                                                      Electric field (V/cm)


                                      Figure 113. The electric field dependence of the drift mobility as a function of the
                                                Tauc gap. The solid curves are fit of the data to equation (1).


                                                                                                                                           101
                     0.29                                                                          12
                            (a)
                     0.28                                                                          10




                                                                              Range of LRPF (µm)
Depth of LRPF (eV)




                     0.27                                                                           8

                     0.26                                                                           6

                     0.25                                                                           4

                     0.24                                                                           2
                                                                                                         (b)
                     0.23                                                                           0
                            1.0           1.1            1.2     1.3    1.4                              1.0         1.1     1.2     1.3   1.4
                                          Tauc gap (eV)                                                              Tauc gap (eV)

                              Figure 114. The depth (a) and range (b) of potential fluctuations
                                              as a function of the Tauc gap.




                                                      0.05

                                                      0.04

                                                      0.03
                                  N ∝ VP /L (a.u.)




                                                      0.02
                                                     0.010
                                  2




                                                     0.005


                                                     0.000
                                                               1.0     1.1                         1.2         1.3         1.4
                                                                       Tauc gap (eV)


                              Figure 115. The relative change in the density of charged defects
                                               as a function of the Tauc gap.




                                                                         102
9. Attempt at finding evidence of the existence of long range potential
   fluctuations in single crystal GeSi alloys

We have employed the photoconductive frequency mixing technique to study the charge
transport properties of HWCVD a-SiGe:H alloys as a function of alloy composition. Evidence
for the presence of long-range potential fluctuations in a-SiGe:H is revealed from the
measurements of electric field dependence of the drift mobility, and the effect of the long-range
potential fluctuations is enhanced by the addition of Ge to the alloy system that results in the
deterioration of the opto-electronic properties of a-SiGe:H. The fact that the monotonic increase
in the depth and the decrease in the range of the potential fluctuations with increasing Ge content
in this alloy system suggests that compositional disorder may also play a role in the long-range
potential fluctuations. Therefore, it is of great interest to examine if this is the case in crystalline
SixGe1-x alloys.

We have obtained a series of single crystal GeSi alloys from RCA Laboratories, David Sarnoff
Research Center, Princeton, New Jersey. The compositions of the GeSi alloys were determined
to 1at.% by X-ray powder pattern measurements of the lattice constants. We performed
photomixing transport measurements on one sample with Ge concentration of 52.3 at.%.
Although we succeeded in preparing ohmic contacts by evaporating Al on the sample surface
followed by alloying in a ~10-6 Torr vacuum at 600°C for 5-10 min, we failed to observe any
electric field dependence of the DC photocurrent and photomixing signals because the
photoconductivity signal was hardly observed, nor was the mixing signal. This is because the
conductivity is confined to a narrow surface layer due to strong absorption (>103 cm-1) under
high photon energy illumination (He-Ne laser, 1.959 eV). In order to measure the photoresponse
and photomixing signal, the sample has to be as thin as a few microns. We anticipate thin film
crystalline GeSi alloys are available, which can be prepared either by mounting thicker pieces on
a glass or sapphire slide with glycol phthalate resin and polishing to size, or by MBE for
example.




                                                 103
10. Photoconductive Frequency Mixing Measurements on TCO

We obtained samples of 1% Cd doped HRT, 1% Zn doped HRT and undoped samples from Randy
Bohn at the University Toledo to see if we can measure the drift mobility by frequency mixing. We
observed a photoconductivity signal at ~5286.5 cm-1 employing our Argon laser but we did not
observe a mixing signal initially; since it was not clear what the spectral range of the absorption
was for these films, we did not employ long integration times. Since that time, Randy Bohn has
supplied us with the spectral absorption of these samples, and we will attempt to measure the drift
mobility as was suggested by Bolko von Roedern.




                                              104
11. Comparison of intrinsic film properties and device performance

During the last period a set of three intrinsic amorphous Si – film samples and related devices
was investigated. The samples were made by S. Guha’s group using the PECVD technique at
United Solar. The n-i-p structures were prepared on stainless steel and ITO-coated. The intrinsic
layers were grown according to the same recipe that was used for the respective films.

The idea is to investigate the relation between

•   the initial a-Si:H film transport properties and the initial performance of the device
•   the decay behavior of both film and related device.

Table 22 gives a summary of the sample preparation as provided by the Uni-Solar group.


Table 22. Growth conditions of the films and the respective i-layers in the n-i-p structures.

Sample ID, film      Sample ID, device         Growth conditions
R8795                R8791                     SiH4 with no dilution, using RF
R8796                R8792                     SiH4 with H2 dilution, using RF
R8794                R8793                     SiH4 with H2 dilution, using VHF



Figures 116 and 117 show data taken from the intrinsic a-Si:H films R8796 and R8794. Both
were produced with hydrogen diluted SiH4. Unfortunately, we were not able to detect any
photocurrent with sample R8795 which was produced without dilution. However, data could be
obtained for all nip structures. The illumination intensity of one sun of the HeNe line was
employed.




                                                  105
                                 12


 Photoconductivity (10 / Ω cm)    9
                                                                                                                    Figure 116. Light-induced
-4




                                                                                                                   changes in photoconductivity,
                                  6                                                                                 mobility, and lifetime of film
                                                                                                                   R8796 (H2 diluted, using RF).
                                  3



                                  0
                                       0.1           1         10            100
                                                   Illumination time (min)

                                 0.8                                                                      60
                                                                                                          55
                                 0.7
                                                                                                          50
                                 0.6                                                                      45
 Mobility (cm / V s)




                                                                                                          40


                                                                                          Lifetime (ns)
                                 0.5
                                                                                                          35
2




                                 0.4                                                                      30
                                                                                                          25
                                 0.3
                                                                                                          20
                                 0.2                                                                      15
                                                                                                          10
                                 0.1
                                                                                                              5
                                 0.0                                                                          0
                                       0.1           1          10           100                                   0.1            1         10            100

                                                   Illumination time (min)                                                      Illumination time (min)


                                 3.5
Photoconductivity (10 / Ω cm)




                                 3.0

                                 2.5
                                                                                                                         Figure 117. Light-induced
                                                                                                                        changes in photoconductivity,
-4




                                 2.0
                                                                                                                         mobility, and lifetime of film
                                 1.5                                                                                      R8794 (H2 diluted, using
                                 1.0                                                                                                VHF).
                                 0.5

                                 0.0
                                             0.1         1           10            100
                                                   Illumination time (min)


                                 0.4                                                                               30

                                                                                                                   25
Mobility (cm / V s)




                                 0.3
                                                                                                   Lifetime (ns)




                                                                                                                   20
2




                                 0.2                                                                               15

                                                                                                                   10
                                 0.1
                                                                                                                    5

                                 0.0
                                             0.1         1           10            100                              0
                                                                                                                          0.1          1          10            100
                                                   Illumination time (min)                                                        Illumination time (min)


                                                                                         106
The difference between the films R8796 and R8794 is the plasma excitation frequency as can be
seen in Table 22. The VHFCVD technique allows for an increase of the deposition rate at an
almost linear function of the frequency to 5-10 times the deposition rate in the case of RF
frequencies. Also, a reduced deposition temperature is possible, both of which is relevant in
device applications. The film quality can still be high due to a decrease of the silicon-ion energy
while the ion-flux is high. The transport parameters, however, are mainly dependent on other
growth parameters which have to be reoptimized as one changes from RF to VHF.

In the case of the present two film samples we find for R8796 (RF excitation) an almost double
as high a mobility as for R8794 (VHF excitation) in the annealed state. Also, the lifetime ratio
between R8794 and R8796 is in the same range. From this we can clearly conclude that the
transport properties of the RF – film is superior. As far as the decay is concerned, we find that
for both samples the decrease of the lifetime is dominant which is also reflected in the results for
the Long-Range Potential Fluctuation range and depth before and after light-soaking. These
quantities are summarized in Table 23. It can be seen that neither range nor depth change much
which shows that the density of charged defects stays more or less constant.

Table 23. Long-range potential fluctuations.

 Sample                State                    LRPF range (nm)        LRPF depth (eV)
                               annealed                34.6                   0.07
       R8796
                           light-soaked                34.5                   0.08
                               annealed                39.8                   0.09
       R8794
                           light-soaked                36.2                   0.09

After 180 minutes of illumination, the decay of the photoconductivity of the RF film is a little
higher in the case of the RF sample than for the VHF film. Since in a wide range H2 dilution both
improves both opto-electronic properties and stability of amorphous silicon films, it is
interesting to see how the device behavior changes with respect to dilution and increase of
frequency.

Figure 118 shows the initial I/V-curves for all nip structures. Again, the nip device made with an
intrinsic film using RF (R8792) shows higher initial short circuit current than R8793 (VHF). The
structure with the intrinsic layer made without dilution shows the lowest initial values. Figure
119 shows the square root of the mixing power for each sample which is linear to the high-
frequency part of the photocurrent. It can be seen that only with back-biasing to –1.5V the
photomixing signals differ noticeably. As already discussed in the 98-99 annual report, the main
difficulty of mixing measurements on devices is the non-uniform profile of the defect
distribution and the electric field. Therefore, interface effects might play a more dominant role
for forward-biasing than the actual intrinsic film properties.




                                                107
                                        0.5
  dc - photoconductivity (10−4 Ω cm )
-1




                                        0.0
-1




                                                                                                             Figure 118.
                                        -0.5                                                           dc-photoconductivity for
                                        -1.0        R8791                                                 nip-devices in the
                                                                                                           annealed state.
                                        -1.5

                                        -2.0     R8793
                                                                        R8792
                                        -2.5

                                        -3.0

                                          -2.0    -1.5   -1.0    -0.5     0.0     0.5   1.0    1.5
                                                                  BIAS (V)



                                        35

                                        30                      R8793
sqrt(mixing power) (fW )
1/2




                                        25
                                                                                                            Figure 119.
                                        20
                                                                                                       Square root of mixing
                                                                                R8792                signal for nip-devices in
                                        15                                                              the annealed state.

                                        10
                                                  R8791
                                         5

                                         0
                                         -2.0     -1.5   -1.0    -0.5     0.0     0.5   1.0    1.5
                                                                  BIAS (V)




                        Figures 120-122 finally give an overview of the light-induced decay measurements on the nip
                        devices. While for sample R8793 (VHF) the back-bias photomixing signal seems to almost
                        mimic the dc – short circuit current, we find the decay of the ac signal to be greater than the dc
                        current for R8792 (RF) and the reverse case for R8791 (undiluted). Assuming that both
                        illumination intensity (about 20mW laser power) and back-biasing is sufficient in order to
                        neutralize charged defect states near the interfaces, the present results would suggest that the
                        decay of the intrinsic film in R8792 contributes less to the over-all decay of the solar cell than in
                        the case of R8793 which would be consistent with the slightly smaller decay of the mobility of
                        the R8796 film (which is incorporated in R8792) than for R8794 even though the over-all decay
                        of the photoconductivity is actually higher for R8796.




                                                                                              108
In summary this would mean that the decay of Isc of the solar cells seems to scale with the initial
performance of the respective single films. It certainly would have been more informative if we
had been able to also acquire decay data on sample R8791 but in any case the experiments
showed that

• the initial cell and film performances for the RF and VHF samples are consistent
• it is not clear whether the light-induced decay of the nip structures scale with the related
   films.
• There is a scaling between the properties of the film and ac- and dc- measurements on
   devices.




                                              1.0                                                                                              1.0


                                                                                             Normalized mixing signal @ 1.25V BIAS
           Normalized short circuit current




                                              0.9                                                                                              0.9


                                              0.8                                                                                              0.8


                                              0.7                                                                                              0.7


                                              0.6                                                                                              0.6


                                                        1               10             100                                                                     1             10             100

                                                            Illumination time (min)                                                                                    Illumination time (min)



                                                    Figure 120. Normalized decay of the dc photocurrent in short circuit condition
                                                                and the back-bias mixing signal in nip device R8791.



                                              1.0                                                                                                        1.0
                                                                                                                Normalized mixing signal @ -1.25V BIAS
  Normalized short circuit current




                                              0.9                                                                                                        0.9


                                              0.8                                                                                                        0.8


                                              0.7                                                                                                        0.7


                                              0.6                                                                                                        0.6

                                                        1              10              100                                                                         1              10             100
                                                             Illumination time (min)                                                                                    Illumination time (min)

                                                    Figure 121. Normalized decay of the dc photocurrent in short circuit condition
                                                                and the back-bias mixing signal in nip device R8792.




                                                                                             109
                                                                                                                           1.0




                                                                                  Normalized mixing signal @ -1.25V BIAS
                                   1.0
normalized short circuit current




                                   0.9                                                                                     0.9



                                   0.8                                                                                     0.8



                                   0.7                                                                                     0.7



                                   0.6                                                                                     0.6


                                         1             10                  100                                                   1         10             100

                                                 Illumination time (min)                                                             Illumination time (min)



                                             Figure 122. Normalized decay of the dc photocurrent in short circuit condition
                                                         and the back-bias mixing signal in nip device R8793.




                                                                                  110
12. Photomixing Measurements on SiGe P-I-N Devices (Supplied by
    Xunming Deng, University of Toledo)


The photomixing measurement method has been shown to be a successful tool in determining
photomixing lifetime and mobility of charge carriers in amorphous and microcrystalline silicon
as well as silicon-germanium and other alloy films which are measured in in-plane direction
employing coplanar contacts. Here, one deals with a well defined uniform electric field due to an
external voltage, a spacially uniform distribution of charged and neutral defects, and a free
charge carrier generation profile perpendicular to the field direction.

In p-i-n structures, in contrast to single films with coplanar ohmic contacts, one usually finds
• a space charge distribution in the intrinsic layer near the p-i and i-n interfaces
• a non-uniform electron-hole generation profile in field direction
• and thus a generally non-uniform electric field distribution across the i-layer

Additionally, the ac behavior of a p-i-n device is determined by the device geometry and the
resulting capacitance. On the other hand, due to the high mixing frequency used, the transient
behavior does not play an important role.

The BIAS dependent photomixing power that one detects for different devices under different
illumination conditions is thus necessarily affected by many more parameters than it is in the
case of single layer films. Even worse, some of the parameters, particularly the electric field
distribution, depends on both the deployed materials and the interfaces in a complicated way that
is still not very well understood. Therefore, instead of incorporating many unknown parameters
into the photomixing model, it seems more advisable to restrict measurements and data analysis
to certain regimes where at least analytical models based on simplified assumptions may be
valid.

The most important analytical approach is probably the uniform-field model by Crandall [29],
whose main assumptions are: uniform field, negligible diffusion in the intrinsic layer, and the use
of the Shockley-Read-Hall model for recombination for a two state recombination center. Later,
Hubin and Shah [30] presented a variation of Crandall’s model introducing three-states
recombination centers. The majority of publications, however, present numerical approaches
which might yield more realistic results but have the drawback of making it hard to draw
immediate physical conclusions. In a recent publication by Asensi et al [31], analytical
expressions for the recombination current and the short-circuit resistance are deduced from
results of numerical simulations.

Dangling bonds within the intrinsic layer near the interfaces are thought to be usually charged
due to the relative shift of the Fermi level to the defect states. Dangling bonds in a charged state,
in turn, have a much higher cross section as recombination centers so that a main part of the total
recombination processes in the i-layer - which represent the main limitation of the p-i-n-cell
efficiency - is assumed to take place near the interfaces. Moreover, these dangling bond states
near the interfaces build up space charge regions which lead to higher electric field strengths
near the interfaces and accordingly lower fields in the center. In fact, the thickness of the i-layer


                                                111
is usually kept as thin as possible, so that even with a certain space charge present, the field near
the center is sufficiently strong to deplete the generated charge carriers.

However, according to Asensi et al., a uniform-field approximation is possible in the special case
of high illumination intensity (Ishort-circuit > 10mA/cm2) and short circuit or back-biasing
condition, i.e. when the charged defect states near the p-i and i-n interfaces are neutralized and
no charge injection into the intrinsic layer takes place. In this situation, with space charge regions
depleted through the strong illumination, also secondary quantities affecting the photomixing
power like the device capacitance should not change dramatically with different bias voltages in
reverse direction.

The photomixing technique, due to the high beating frequency, merely yields a picture of
shallow trap states near the conduction band and is insensitive to deep level defects. Therefore,
only for the above mentioned conditions it seems possible to obtain information about the
transport properties of the intrinsic layer without the need to deconvolute a complex amount of
parameters most of which are not even known. On the other hand, this might enable us to keep
track of a certain material-related aspect, i.e. the charge carrier mobility, which is otherwise
somewhat lost once single films are deposited within a multilayer structure.

Figures 123 shows the square root of the photomixing power under back-biasing.



                    8                                                    30

                    7                                                    25                            high light intensity
photocurrent (mA)




                    6                                                    20
                                                         sqrt(Pmixing)




                                                                               0
                    5                                                    15
                                                                               -2
                                                                               -4                              V0 = 0.79V
                    4                                                    10    -6
                                                                               -8

                    3                                                    5    -10
                                                                              -12
                                                                              -14
                    2                                                    0          -3   -2   -1   0

                        -3     -2        -1    0     1                    -4.5-4.0-3.5-3.0-2.5-2.0-1.5-1.0-0.5 0.0 0.5 1.0 1.5

                                    BIAS (V)                                                       BIAS (V)


                        Figure 123. DC photocurrent and the square root of the photomixing power
                                            (inset: derivative of sqrt(Pmix)).


While the dc photocurrent shows expected saturation behavior when the p-i-n device is reverse-
biased, in the case of the ac current (as represented by the square root of the mixing power) we
find (at least) two distinct regimes, a low-field and a high-field regime.
In the high-field regime, we find an increase with a saturation tendency at sufficiently high
fields. The mobility of electrons in a p-i-n structure when approaching high fields has been



                                                           112
             described in terms of field-assisted thermal release of electrons from traps within the multiple
             trapping model [32].

             However, for biases below the point of inflection, the shape of the curve rather resembles the
             field-dependent mobility which is also observed in thin film samples with coplanar contacts. The
             dotted curve which is fit into the lower-field part of the photomixing current curve points out
             where the photomixing signal deviates from it’s initial shape. According to this fit, the
             photomixing signal would become zero at about +0.8V, which would mean that at this point the
             built-in field zero-crosses.


                                                                                                         0
             25                                              -2
                                     ILaser = 180mWcm
                                                                                                         -2
             20
                                                                                                         -4

                                                                               normalized sqrt(P mix)
                                         -2
             15             90mWcm
sqrt(Pmix)




                                                                                                         -6

             10                                                                                          -8
                                    -2
                       50mWcm
                                                                                                        -10
              5

                                                                                                        -12
              0
                                                                                                        -14
              -3.5   -3.0    -2.5   -2.0      -1.5   -1.0   -0.5   0.0   0.5                              -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5   0.0   0.5   1.0
                                         BIAS (V)                                                                             BIAS (V)

             Figure 124. Light intensity dependent photomixing current. Right hand-side: The derivatives
                   show a shift of the point of inflection to higher biases at higher light intensities.


             However, this fitting only works for high illumination levels. For lower light intensities we
             observe a shift towards higher electric fields (Figure 124). It is noteworthy, however, that even
             for (very) high illumination levels a uniform field within the intrinsic layer is at best a good
             approximation. For lower levels, this simplified model is certainly not feasible. Absolute values
             for the mobility across the intrinsic layer seem therefore – in contrast to single films – not very
             meaningful. However, investigating the evolution of the bias-dependent photomixing curves
             under light-soaking provide a separation of effects due to geometric or initial material properties
             from those due to light-induced defects as the following figures show.




                                                                                                        113
                         after 1min illumination
                    6                                                                                          36
                                                                                                               32
                    5
                                                                                                               28
                    4                                                                                          24
Photocurrent(mA)



                                         BIAS (V)                                                                                   BIAS (V)                                                                                     1/2
                          -4        -2              0         2                                                      -4        -2              0              2
                                                                                                                                                                                                    -1.21V, 14 fW
                                                                  1                                            20                                                 8




                                                                                              sqrt(Pmixing)
                    3                                                                                                                                             6
                                                                  0                                                                                               4
                                                                                                               16                                                 2
                                                                                                                                                                  0
                    2




                                                                                                                                                                        d sqrt(P m ixing) / d V
                                                                  -1
                                                                                                               12                                                 -2
                                                                                                                                                                  -4




                                                                       dIpho to/dV
                                                                  -2                                                                                              -6
                    1                                                                                          8                                                  -8
                                                                  -3                                                                                              -10
                                                                                                               4                                                  -12
                                                                  -4                                                                                              -14
                    0                                                                                                                                             -16
                                                                  -5                                           0                                                  -18
                                                                                                                                                                  -20
                    -1                                                                                         -4
                               -3         -2             -1                    0      1   2                      -7       -6        -5        -4        -3        -2                                     -1        0       1         2
                                                        BIAS (V)                                                                                   BIAS (V)

                         after 16min illumination
                     6
                                                                                                               32
                     5
                                                                                                               28

                     4                                                                                         24                                                                                                              1/2
 Photocurrent(mA)




                                         BIAS (V)                                                                                   BIAS (V)                                                      -1.5V, 15 fW
                          -4        -2              0         2                                                      -4        -2              0              2
                                                                                                               20                                                 4
                                                                                               sqrt(Pmixing)




                     3                                            0

                                                                                                               16                                                 0
                                                                  -1
                     2                                                                                                         -1.6V




                                                                                                                                                                               d sqrt(P m ixing) / d V
                                                                                                               12                                                 -4
                                                                  -2
                                                                        dIpho to/dV




                     1                                            -3                                            8                                                 -8

                                                                  -4                                                                                              -12
                     0                                                                                          4
                                                                  -5                                                                                              -16
                                                                                                                0
                    -1
                               -3         -2             -1                      0    1   2                         -7    -6        -5        -4        -3         -2                                    -1        0       1         2
                                                        BIAS (V)                                                                                   BIAS (V)

                         after 440min illumination
                     5
                                                                                                               16

                     4
                                                                                                               12               BIAS (V)
 Photocurrent(mA)




                                         BIAS (V)
                     3    -4        -2              0         2                                                      -4   -3   -2        -1    0        1     2
                                                                                                                                                               2
                                                                                              sqrt(Pmixing)




                                                                  0
                                                                                                                8                                                 0
                     2
                                                                                                                                                                        d sqrt(Pmixing) / d V




                                                                  -1
                                                                                                                                                                  -2
                                                                        dIpho to/dV




                     1                                            -2
                                                                                                                4
                                                                                                                                                                  -4
                                                                  -3
                     0                                                                                                                                            -6
                                                                  -4                                            0
                                                                                                                                                                  -8
                    -1
                               -3         -2             -1                      0    1   2                         -8    -7    -6        -5       -4        -3              -2                               -1       0   1         2
                                                        BIAS (V)                                                                                   BIAS (V)

                                Figure 125. dc and mixing currents after different illumination times
                                                     (Sample Toledo-GD112).



                                                                                              114
.
                           after 5min illumination
                                                                                                                                      12
                      5


                      4
                                                                                                                                      8                                                                                        -0.9V, 5.2 fW
                                                                                                                                                                                                                                               1/2
 Photocurrent(mA)




                                                 BIAS (V)                                                                                                       BIAS (V)
                      3      -6        -4         -2            0             2                                                                 -4              -2             0           2
                                                                                                                                                                                               2




                                                                                                                      sqrt(Pmixing)
                                                                                      0

                      2                                                               -1                                                                                                       0
                                                                                                                                      4




                                                                                                                                                                                                      d sqrt(Pmixing) / d V
                                                                                      -2

                                                                                           dIp hoto/dV
                      1                                                                                                                                                                        -2
                                                                                      -3

                                                                                      -4                                                                                                       -4
                      0                                                                                                               0
                                                                                      -5
                                                                                                                                                                                               -6
                      -1
                                 -5        -4          -3            -2            -1                    0   1   2                        -8         -7     -6        -5       -4    -3            -2                          -1    0    1          2
                                                                    BIAS (V)                                                                                                       BIAS (V)
                           after 100min illumination
                                                                                                                                     12
                      5


                      4
                                                BIAS (V)                                                                              8                     BIAS (V)                                                          -1.31V, 5.3 fW
                                                                                                                                                                                                                                               1/2
   Photocurrent(mA)




                            -6        -4          -2        0             2                                                                -6         -4         -2        0         2
                      3
                                                                                                                     sqrt(Pmixing)




                                                                                   0                                                                                                       6

                                                                                                                                                                                           4




                                                                                                                                                                                                     d sqrt(Pmixi ng) / d V
                      2                                                            -1                                                 4                                                    2
                                                                                           dIphoto/dV




                                                                                   -2                                                                                                      0
                      1                                                                                                                                                                    -2
                                                                                   -3
                                                                                                                                      0                                                    -4
                      0                                                            -4
                                                                                                                                                                                           -6

                             -5            -4          -3           -2            -1                     0   1   2                        -8     -7        -6         -5   -4        -3         -2                             -1   0    1           2
                                                                BIAS (V)                                                                                                           BIAS (V)
                          after 425min illumination
                      5
                                                                                                                                      8
                                                                                                                                                                                                                                               1/2
                      4                                                                                                                                                                                                       -1.55V, 5.4 fW
                                                                                                                                                           BIAS (V)
Photocurrent(mA)




                                                BIAS (V)
                            -6        -4         -2         0             2                                                                     -4              -2         0              2
                      3                                                                                                                                                                    2
                                                                                                                      sqrt(Pmixing)




                                                                                  0
                                                                                                                                      4
                                                                                                                                                                                          0
                                                                                                                                                                                                d sqrt(Pmixi ng) / d V




                                                                                  -1
                      2
                                                                                       dIph oto/dV




                                                                                  -2                                                                                                      -2
                      1
                                                                                  -3
                                                                                                                                      0                                                   -4

                      0                                                           -4

                                                                                                                                                                                          -6
                            -5         -4             -3            -2            -1                     0   1   2                    -8        -7         -6        -5    -4       -3         -2                             -1    0    1           2
                                                                BIAS (V)                                                                                                       BIAS (V)


                                                Figure 126. dc and mixing currents after different illumination times
                                                                     (Sample Toledo-GD111).



                                                                                                                                      115
                            after 1min illumination
                                                                                                                                                        10

                        4
                                                                                                                                                           8

                                                                                                                                                                                BIAS (V)
Photocurrent(mA)



                        3                        BIAS (V)
                                                                                                                                                           6         -5   -4        -3    -2         -1     0    1
                            -6          -4             -2         0             2
                                                                                                                                                                                                                  2




                                                                                                                                        sqrt(Pmixing)
                                                                                                                                                           4                                                      1
                        2                                                           0

                                                                                                                                                                                                                  0




                                                                                                                                                                                                                         d sqrt(Pmixi ng) / dV
                                                                                         dIph oto/dV
                                                                                                                                                           2                                                      -1
                        1                                                           -1
                                                                                                                                                                                                                  -2
                                                                                                                                                           0
                                                                                                                                                                                                                  -3
                        0
                                                                                    -2                                                                                                                            -4
                                                                                                                                                        -2
                        -6             -5          -4        -3            -2                      -1               0       1       2                    -10 -9                -8        -7     -6         -5    -4               -3                             -2   -1   0   1
                                                                  BIAS (V)                                                                                                                                BIAS (V)
                            after 425min illumination

                        4
                                                                                                                                                           8

                        3                                                                                                                                                       BIAS (V)
Photocurrent(mA)




                                                  BIAS (V)
                             -6             -4          -2        0             2                                                                               -6             -4                -2              0
                                                                                                                                                                                                                  2
                                                                                                                                           sqrt(Pmixing)




                        2                                                                                                                                  4
                                                                                    0




                                                                                                                                                                                                                       d sqrt(Pmixi ng) / d V
                                                                                                                                                                                                                  0
                                                                                         dIph oto/dV




                        1
                                                                                                                                                                                                                  -2
                                                                                    -1
                                                                                                                                                           0
                        0
                                                                                                                                                                                                                  -4
                                  -5             -4          -3            -2                           -1              0       1                          -10 -9              -8        -7     -6         -5    -4               -3                             -2   -1   0   1
                                                                  BIAS (V)                                                                                                                                BIAS (V)


                            after 3min illumination
                        5                                                                                                                                  10


                        4                                                                                                                                    8
                                                      BIAS (V)                                                                                                                       BIAS (V)
     Photocurrent(mA)




                                 -2.0 -1.5 -1.0 -0.5 0.0               0.5           1.0                                                                              -2.0      -1.5          -1.0        -0.5   0.0
                                                                                                                                                                                                                   15
                                                                                                                                                             6
                                                                                                                                           sqrt(Pmixing)




                        3                                                                   0
                                                                                                                                                                                                                      10
                                                                                            -1
                                                                                                                                                                                                                                        d sqrt(Pmixi ng) / d V




                                                                                                                                                             4                                                        5
                        2
                                                                                                       dIphoto/dV




                                                                                            -2
                                                                                                                                                                                                                      0
                                                                                            -3                                                               2                                                        -5
                        1
                                                                                            -4                                                                                                                        -10
                                                                                                                                                             0
                                                                                            -5                                                                                                                        -15
                        0
                                            -2                        -1                                            0               1                           -5             -4                -3              -2                                              -1        0       1
                                                                      BIAS (V)                                                                                                                            BIAS (V)



                                                                                                  Figure 127. Sample Toledo-GD110.



                                                                                                                                              116
                          after 300min illumination
                                                                                                           12
                      5


                      4                                                                                     8
                                        BIAS (V)                                                                              BIAS (V)
   Photocurrent(mA)




                          -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0                                             -2.0   -1.5     -1.0        -0.5   0.0




                                                                                           sqrt(Pmixing)
                      3                                          1
                                                                                                                                                        5
                                                                 0




                                                                                                                                                              d sqrt(Pmixi ng) / d V
                                                                 -1                                         4
                      2                                                                                                                                 0


                                                                      dIphoto/dV
                                                                 -2
                                                                 -3                                                                                     -5
                      1
                                                                 -4
                                                                                                            0
                                                                 -5                                                                                     -10
                      0
                                 -2                   -1                           0   1                        -5       -4            -3          -2                                  -1   0   1
                                                     BIAS (V)                                                                                BIAS (V)

                                 Figure 128. dc and mixing currents after different illumination times
                                                      (Sample Toledo-GD109).

Figures 125-128 show data obtained from light-soaking measurements on p-i-n-samples supplied
by the Toledo group. The insets contain the neighbor-averaged derivatives of their parent curves
in order to facilitate the observation of the change of the curve shape during light-soaking. A
summary of the respective I-V data can be found in Table 24.

Table 24. I-V data for a-SiGe solar cells with different i-layers (~2000Å).

                                                                      Jph from QE
                      Sample                Voc (V)                                    FF                                       Pmax (mW/cm2)
                                                                      (mA/cm2)
 GD109
                                            0.681                     19.3             0.530                                    6.97
 Standard
 GD110
                                            0.634                     21.3             0.473                                    6.39
 +15% Ge
 GD111
                                            0.714                     20.0             0.527                                    7.53
 -15% Ge
 GD112
 2.5x H                                     0.813                     18.2             0.486                                    7.19
 dilution


Sample GD112 shows both the highest mixing signals and the largest decay under light-soaking.
However, according to data from Toledo (Table 24) this sample’s filling factor is the second
worst of all samples. It is obvious that the dc p-i-n cell performance and the photomixing signal,
even though associated with the charge carrier mobility, do not directly scale. This was not to be
expected, though. It is clear that the actual device impedance has a great impact on the
photomixing signal power. A comparison of the photomixing curves for two different spots on
the same sample (GD112) as shown in Figure 129 stresses this point.


                                                                                              117
             40
                                                  At one spot (‘spot’ means one separate p-i-n-
                                                  structure among others on the same stainless steel
      30                                          substrate) the curve reaches saturation behavior
                                                  earlier but at lower mixing signals than it does at
                                                  some other spot. A steeper curve at lower fields
sqrt(Pmix)




      20
                                                  and consequently earlier onset of saturation would
                                                  mean higher mobility while the variation in mixing
      10
                                                  power is rather due to the differences in the sample
                                                  impedance which is apparently quite sensitive to
       0                                          fabrication variations. In contrast to this effect the
            -3     -2      -1      0     1
                                                  changes during light soaking (Figure 125), i.e.
                       BIAS (V)
                                                  with parameters other than the defect density kept
                                                  constant, suggest that the charged defect related
          Figure 129. BIAS-dependent ac –         field-dependent mobility regime extends to higher
         photocurrent for two different spots.    electric fields across the i-layer at the saturation
                                                  regime’s expense. After seven hours of light
    exposure, the saturation regime vanishes completely which means that the applied bias has
    become insufficient in order for the degraded sample to enter this region. It is also noteworthy
    that the point of inflection, as marked within the first two viewgraphs of Figure 125, seems to
    more or less remain at the same height while wandering towards higher biases. After 425
    minutes, the photomixing current just reaches the 16 fW1/2 level from which the saturation
    regime evolved in the annealed state. Note also that the dc-current curve seems less affected than
    the ac-curve. In all samples except for GD110, which turns out be the most stable one, we find
    the slope of the ac-curve within the lower-field regime to decrease with illumination time.
    Unfortunately, in most cases we had to limit the maximum reverse bias in order to protect the
    samples from avalanching. On the other hand, former measurements on a-Si samples showed
    saturation much more clearly - probably because the electron mobility is generally higher in a-Si
    than in a-SiGe.

    In order to obtain at least approximate values for the electron mobility in p-i-n-structures the
    photoconductive frequency mixing model has to be adapted appropriately. In particular, the
    capacitance and geometry contributions will have to be treated in such a way that the ac-signal
    dependent transport expressions become independent of those. Therefore, accompanying ac-
    measurements are needed. In a next step, the photomixing transport equations have to be applied
    and tested against results from further measurements on a series of n-i-p-solar cells supplied by
    the Toledo group.

    We also plan a series of experiments on samples in coplanar configuration and p-i-n-devices that
    are prepared under the same growth condition.

    As far as the experiment is concerned, a higher accuracy, particularly of the ac-current curve
    derivative, is desirable. One possible way to obtain more accurate results is a more direct
    measurement of the derivative of the mixing power. Initial experiments employing low-
    frequency bias sweep and log-in amplification techniques are promising. However, there is
    always a certain trade-off between measurement accuracy and the undesirable light-soaking
    effect during the acquisition of a photomixing signal curve.



                                                    118
13. Photo-emission in Air
I.    Statement of the problem


        The broad objectives of this research are to develop optical stimulated photo-emission in
air as a non-destructive technique to characterize semiconductor surfaces. In most applications of
photo-emission the photoelectrons are usually collected in vacuum. However the emitted and
subsequently scattered electrons can be collected across a sufficiently small air gap by a biased
collector and provide information about the electronic structure, composition and chemistry of
the surface. A powerful array of electron spectroscopes exist for detecting chemical impurities on
surface but usually require an ultra-high vacuum environment and are not readily adaptable to
analytic techniques ultimately be used on the production line. These surface analytic techniques
include the scanning Auger Microscope, Secondary Ion Mass Spectroscopy and Electron
Spectroscopy for Chemical Analysis (ESCA). The development of a technique of optical photo-
emission in air will offer a solution to the problem of monitoring surface contamination under
normal factory ambient conditions. In the present proposal, we will develop the technique of
photo-emission in air and explore its use to study a number of interface problems in
semiconductors. These include oxides on semiconductors and homogeneity of composition of
semiconductor alloys.

II.   Implementation of photo-emission in Air


         Figure (130) is a simplified schematic representation of the process of photo-emission in
air. UV light passes through the collector grid to the sample. The bias between the sample and
the collector drives the emitted electrons by a diffusion process. Figure (131) shows the energy
diagrams for two metals with an external voltage, V, in vacuum if light is radiated on the
cathode. Figure (131a) shows the anode reversed biased, i.e. in the retarded field mode; Figure
(131b) shows the space charge limited regime, while figure (131c) delineates the saturation
regime. Figure (132) shows the photo-current versus V for the conditions shown in figure (131).
For large negative values of V (Figure 131a) no current flows since the barrier is too high. At Vc
the emitted electrons have zero velocity. As V becomes more positive, the more energetic
electrons are able to surmount the barrier causing the electron could next to the cathode as per
(Figure 132). This is the space charge limited current and current is given by Child-Langmuir
law.
If the initial photo-emissive is overlaid with a thin film, attenuation in the film will determine the
number of photoelectron collected by the biased collector; on the other hand if the film is also
photo-emissive the collector current can increase.
The structure shown in Figure (130) will be implemented by constructing photo-emissive-
collector probes using Hg vapor UV sources and laser lines from harmonic generators and dye
laser lines; the collectors will be moved over a surface in two dimensions by stepping motors.




                                                 119
III. Results


The results represent were obtained using the Hg vapor UV lamp as a light source. The initial
studies were on the homogeneity of the a-Si:H, interface studies of TCO on the a-Si:H samples
obtained from Qi Wang of NREL.
a- Figure (133), present the results of a sample of a-Si:H and ITO both prepared on stainless
    steel substrate. We found that there is no photo-emission signal from ITO however high yield
    from a-Si:H.
b- Figure (134), shows the photo-emission from a surface with the spots of ITO on a-Si:H. It
    should be noted that the region between the spots shown a uniform high yield from a-Si:H
    layer. However, the region of the ITO spots show a decrease of the photo-emission yield
    relative to the a-Si:H layer. The fact of the photo-emission decrease under the ITO layer
    relative to the a-Si:H layer but it is a positive quantity indicates the UV line passes through
    the ITO layer and generates the photo-emission electrons which penetrate the ITO layer to be
    collected.
c- Figure (135), shows the results of the structure made up of ITO/pin/SS layers. It should be
    observe the decrease of the photo-emission under the ITO layer similar to fig.134.
d- Figure (136), Shows the results of the sample thin Pd/ni/SS. It should be observe that similar
    results as of the ITO layer but the yield is less. This indicate the UV light is partially
    reflected from the Pd layer or the photo-emission yield is less in the case of the Pt/a-Si:H
    layer.
e- Figure (137), shown the results of a-Si:H on c-Si substrate which indicate the relative
    homogeneity of a-Si:H layer.
f- Figure (138), shows the results of the 16 pads which are a-Si:H to µc-Si:H and a-SiGe:H to
    µc-SiGe:H on stainless steel substrate. It should be observe that are differences in the work
    function between the materials. In the present experimental arrangement we use the whole
    spectrum of the Hg lamp. However, the excellent signal to noise in the figure indicates that in
    the new photo-emission setup we plan to build which will employ a known spectrum line and
    variable bias, it will be possible to measure the work function as a function of the various
    pads.
g- Figure (139 a,b,c&d), shows the results of a-SiC:H samples on stainless steel with different
    carbon concentrations(0-40). It is found that the surface of the samples are not homogeneous.




                                               120
Figure 130. Schematic diagram of Photo-emission in air




                         121
  Figure 131. Electron energies between the anode and cathode in a diode under bias
conditions. (a) Anode reversed bias (retarding field); (b) emission of space charge limited
 barrier results from space- charge just outside cathode surface; (c) saturation emission




                                           122
Figure 132. Photo-emission current vs voltage bias




                       123
Figure 133. The photo-emission intensity of ITO and a-Si:H on stainless steel substrate




                                         124
Figure 134. The photo-emission intensity of ITO on a-Si:H on stainless steel substrate.




                                         125
Figure 135. The photo-emission intensity of ITO/pin/SS layers.




                             126
Figure 136. The photo-emission intensity of Pd/ni/SS layers.




                           127
Figure 137. The photo-emission intensity of a-Si:H on c-Si substrate.




                                128
Figure 138. The photo-emission intensity the 16 pads which are a-Si:H to µc-Si:H and a-
                   SiGe:H to µc-SiGe:H on stainless steel substrate.




                                         129
Figure 139(a). The photo-emission intensity of a-SiC:H on stainless steel substrate.




                                        130
Figure 139(b). The photo-emission intensity of a-SiC:H on stainless steel substrate.




                                        131
Figure 139(c). The photo-emission intensity of a-SiC:H on stainless steel substrate.




                                        132
Figure 139(d). The photo-emission intensity of a-SiC:H on stainless steel substrate.




                                        133
14. Subcontract Supported Publications

1. A. Kattwinkel, R. Braunstein and Q. Wang,” Transition from hydrogenated amorphous
   silicon to microcrystalline silicon” MRS Symp.Proc. Vol. 557(Material Research Society
   Spring Meeting (1999)543.

2. J. Liebe, A. Kattwinkel, K. Baerner, G. Sun, S. Dong and R. Braunstein, “ Determination of
   the gap density differences in hydrogenated amorphous silicon and Si/Ge” Material Science
   and Engineering A282(2000) 158-163

3. S.R. Sheng, R. Braunstein and V.L. Dalal, “Electronic and optical properties of high quality
   low bandgap amorphous (Ge, Si) alloys”, MRS Symp.Proc. Vol. 664 (Material Research
   Science Spring Meeting 2001)A8.4.1.

4. S.R. Sheng, R. Braunstein, B.P. Nelson, and Y.Q. Xu ,”Electronic transport study of high
   deposition rate HWCVD a-Si:H by the microwave photomixing technique”, MRS
   Symp.Proc. Vol. 664 (Material Research Science Spring Meeting 2001)A23.4.1.

5. S.R. Sheng , G.S. Sun , J. Liebe , A. Kattwinkel, R. Braunstein, B.P. Nelson, B. von Roedern,
   K. Baerner ,”Electronic properties of hydrogenated amorphous silicon-germanium alloys and
   long-range potential fluctuations”, Material Science and Engineering A325 (2002)490.

6. S.R. Sheng, M. Boshta, R. Braunstein , and V.L. Dalal, “On the electronic transport
   properties of amorphous (Si, Ge) alloys: charged scattering centers and compositional
   disorder”, J.Non-Cryst.Solids 303(2002)201.

7. M. Boshta, R. Braunstein, and G. Ganguly, “Electronic properties of low band gap a-SiGe:H
   alloys prepared by PECVD technique”, accepted for publication in J. Non-crystalline Solids.




                                              134
15. References

[1] E.R. Geissinger, R. Braunstein, S. Dong, and R. Martin, J. Appl. Phys. 69, 1469 (1991).
[2] Y. Tang, R. Braunstein, and B. von Roedern, Appl. Phys. Lett. 63, 2393 (1992).
[3] Y. Tang, R. Braunstein, B. von Roedern, and F.R. Shapiro. Mat. Res. Soc. Symp. Proc. 297,
    407 (1993).
[4] Y. Tang and R. Braunstein, Appl. Phys. Lett. 66, 721 (1995).
[5] Y. Tang and R. Braunstein, J. Appl. Phys. 79, 850 (1996).
[6] Y. Tang, S. Dong, R. Braunstein, and B. von Roedern, Appl. Phys. Lett. 68, 640 (1996).
[7] D. V. Tsu, B. S. Chao, S. R. Ovshinsky, S. Guha, and J. Yang, Appl. Phys. Lett., 71, 1317
    (1997)
[8] S. Sheng, X. Liao, G. Kong, and H. Han, Appl. Phys. Lett., 73, 336 (1998)
[9] S. Sheng, X. Liao, Z. Ma, G. Yue, Y. Wan, and G. Kong, MRS Symp. Proc. 507 (1998)
    p.969.
[10] Meier, P. Torres, R. Platz, S. Dubail, U. Kroll, J.A. Anna Selvan, N. Pellton Vaucher, Ch.
    Hof, D. Fischer, H. Keppner, A. Sha, K. D. Lifert, P. Giannulés, J. Koehler, MRS Symp.
    Proc. 420 (1996).
[11] H. Wiesmann, A. K. Ghosh, T. McMahon, and M. Strongin, J. Appl. Phys. 50, 3752 (1979).
[12] H. Matsumera. Jpn. J Appl. Phys., Part 2, 25, L949 (1986).
[13] J. Cifre, J. Bertomeu, J. Puigdollers, M.C. Polo, J. Andreu, and A. Lioret, Appl. Phys.A:
     Solids Surf. 59, 645 (1994).
[14] A. R. Middya, J. Guillet, J. Perrin, A. Lioret, and J.E. Boirree in Proceedings of the 13th
     European Photovoltaic Solar Energy Conference, Nice 1995 edited by W. Freiesleben, W.
     Palz, H.A. Ossenbrink and P. Helm (H.S. Stephens & Associates, Bedford, U. K., 1995). p. 3.
[15] M. Heintze, R. Zedlitz, H.N.Wanka, and M. B.Schubert, J. Appl. Phys. 79, 2699 (1996).
[16] F. Diehl, W. Herbst, B. Schröder, H. Oechsner, MRS Symp. Proc. 467 (1997), p.451.
[17] R. Brüggemann, C Main Phys. Rev. D 57, R15080 (1998).
[18] A. A. Langford, M. L. Fleet, B. P. Nelson, W. A. Lanford, and N. Maley, Phys. Rev. B 45,
     13367 (1992).
[19] S. Guha, J. Yang, S. Jones, Y. Chen, and D. Williams, Appl. Phys. Lett. 61,144 (1992).
[20] T. Takagi, R. Hayashi, A. Payne, W. Futako, T. Nishimoto, M. Takai, M. Kondo and A.
     Matsuda, MRS Symp. Proc. 557, 105 (1999).
[21] Brent P. Nelson, Yueqin Xu, A. Harv Mahan, D.L. Williamson and R.S. Crandall, MRS
     Symp. Proc. 609 (in press).
[22] A.V. Gelatos, K. K. Mahadavi, and J. D. Cohen, J. P. Harbison, Appl. Phys. Lett. 53 403
     (1988).
[23] S.R. Sheng, G.S. Sun, J. Liebe, A. Kattwinkel, R. Braunstein, B.P. Nelson and B. von
     Roedern, (submitted to Materials Science and Engineering A).
[24] S.J. Jones, Y. Chen, D.L. Williamson, R. Zedlitz and G. Bauer, Appl. Phys. Lett. 62, 3267
     (1993).
[25] D. Fink and R. Braunstein, Phys. Stat. Sol. (b) 73, 361 (1976).
[26] V.L. Dalal and Z.Y. Zhou, MRS Symp. Proc. 609 (in press).
[27] G.D. Cody, T. Tiedje, B. Abeles, B. Brooks, and Y. Goldstein, Phys. Rev. Lett. 47 1480 (1981).
[28] D. Kurnia, R.P. Barclay, and J.M. Boud, J. Non-Cryst. Solids 137&138, 375 (1991).
[29] R.S. Crandall, J. Appl. Phys. 54, 7176 (1983).


                                              135
[30] J. Hubin and A. V. Shah, Philos. Mag. B 72, 589 (1995).
[31] J. M. Alensi, J. Merten, C. Voz, and J. Andreu, J. Appl. Phys. 85, 2939 (1999).
[32] G. Juska, K. Arlauskas, J. Kocka, M. Hoheisel, and P. Chabloz, Phys. Rev. Let 75, 2984
(1995).




                                           136
                                                                                                                                                       Form Approved
REPORT DOCUMENTATION PAGE                                                                                                                            OMB NO. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,
gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this
collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Leave blank)                2. REPORT DATE                               3. REPORT TYPE AND DATES COVERED
                                                   December 2002                                Final Subcontract Report
                                                                                                20 April 1998–30 June 2002
4. TITLE AND SUBTITLE
                                                                                                                                         5. FUNDING NUMBERS
   Photocharge Transport and Recombination Measurements in Amorphous Silicon Films                                                           CF: XAK-8-17619-24
   and Solar Cells by Photoconductive Frequency Mixing: Final Subcontract Report,                                                            PVP3.5001
   20 April 1998–30 June 2002
6. AUTHOR(S)
   R. Braunstein, M. Boshta, S. Sheng, A. Kattwinkel, J. Liebe, and G. Sun
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)                                                                                       8. PERFORMING ORGANIZATION
  University of California                                                                                                                  REPORT NUMBER
  405 Hilgard Avenue
  Los Angeles, California 90024-1406
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)                                                                                  10. SPONSORING/MONITORING
   National Renewable Energy Laboratory                                                                                                      AGENCY REPORT NUMBER
   1617 Cole Blvd.
   Golden, CO 80401-3393                                                                                                                        NREL/SR-520-33173

11. SUPPLEMENTARY NOTES
   NREL Technical Monitor: B. von Roedern
12a.    DISTRIBUTION/AVAILABILITY STATEMENT                                                                                              12b.    DISTRIBUTION CODE
        National Technical Information Service
        U.S. Department of Commerce
        5285 Port Royal Road
        Springfield, VA 22161
13. ABSTRACT (Maximum 200 words) The tasks carried out under this subcontract focused on characterizing the charge transport,
opto-electronic, and structural properties of a number of amorphous and microcrystalline semiconductors prepared by several
techniques. The dominant approach to accomplish the tasks of the present phase of the program is the photoconductive
frequency mixing technique. This technique enabled us to determine separately the drift mobility and the photomixing lifetime
of the photogenerated carriers. The technique is based on the idea of heterodyne detection for photoconductors. When two
similarly polarized monochromatic optical beams of slightly different frequencies are incident on a photoconductor, the
photocurrent produced, when a dc bias is applied, will contain components resulting from the square of the sum of the incident
electric fields. Consequently, a photocurrent composed of a dc and a microwave current due to the beat frequency of the
incident fields will be produced; these two currents allow a separate determination of the drift mobility and the photomixing
lifetime. In the present work, we improved the instrumentation of the photomixing measurements by applying bias pulses of
arbitrary width and frequency. The longitudinal modes of a He-Ne laser were used to generate a beat frequency of 252 MHz;
all the measurements were performed at this frequency for the data indicated in the accompanying figures and tables. Results
from this technique, as well as FTIR, XRD, SAXS, and optical spectroscopy, are presented in the full report.
                                                                                                                                         15. NUMBER OF PAGES
14. SUBJECT TERMS:   PV; photomixing; metastability; photocharge transport; amorphous
silicon films; photoconductive frequency mixing; optical spectroscopy; drift mobility;
photomixing lifetime                                                                                                                     16. PRICE CODE


17. SECURITY CLASSIFICATION                     18. SECURITY CLASSIFICATION                  19. SECURITY CLASSIFICATION                 20. LIMITATION OF ABSTRACT
    OF REPORT                                       OF THIS PAGE                                 OF ABSTRACT
       Unclassified                                   Unclassified                                Unclassified                                  UL

NSN 7540-01-280-5500                                                                                                                             Standard Form 298 (Rev. 2-89)
                                                                                                                                                             Prescribed by ANSI Std. Z39-18
                                                                                                                                                                                   298-102

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:6
posted:10/4/2011
language:English
pages:151